id	sid	tid	token	lemma	pos
cana-5267	1	1	threatfinderai	threatfinderai	NOUN
cana-5267	1	2	:	:	PUNCT
cana-5267	1	3	automated	automate	VERB
cana-5267	1	4	threat	threat	NOUN
cana-5267	1	5	modeling	modeling	NOUN
cana-5267	1	6	applied	apply	VERB
cana-5267	1	7	to	to	ADP
cana-5267	1	8	llm	llm	NOUN
cana-5267	1	9	system	system	NOUN
cana-5267	1	10	integration	integration	NOUN
cana-5267	1	11	communications	communication	NOUN
cana-5267	1	12	on	on	ADP
cana-5267	1	13	applied	apply	VERB
cana-5267	1	14	nonlinear	nonlinear	ADJ
cana-5267	1	15	analysis	analysis	NOUN
cana-5267	1	16	issn	issn	NOUN
cana-5267	1	17	:	:	PUNCT
cana-5267	1	18	1074	1074	NUM
cana-5267	1	19	-	-	PUNCT
cana-5267	1	20	133x	133x	NUM
cana-5267	1	21	vol	vol	VERB
cana-5267	1	22	32	32	NUM
cana-5267	1	23	no	no	NOUN
cana-5267	1	24	.	.	PUNCT
cana-5267	2	1	10s	10	NOUN
cana-5267	2	2	(	(	PUNCT
cana-5267	2	3	2025	2025	NUM
cana-5267	2	4	)	)	PUNCT
cana-5267	2	5	1658	1658	NUM
cana-5267	2	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5267	2	7	athreat	athreat	NOUN
cana-5267	2	8	modeling	model	VERB
cana-5267	2	9	framework	framework	NOUN
cana-5267	2	10	for	for	ADP
cana-5267	2	11	llm	llm	NOUN
cana-5267	2	12	system	system	NOUN
cana-5267	2	13	integration	integration	NOUN
cana-5267	2	14	leveraging	leverage	VERB
cana-5267	2	15	nlp	nlp	NOUN
cana-5267	2	16	and	and	CCONJ
cana-5267	2	17	machine	machine	NOUN
cana-5267	2	18	learning	learn	VERB
cana-5267	2	19	1basavraj	1basavraj	NUM
cana-5267	2	20	gadade	gadade	NOUN
cana-5267	2	21	,	,	PUNCT
cana-5267	2	22	2prof	2prof	NUM
cana-5267	2	23	.	.	PUNCT
cana-5267	3	1	g.	g.	PROPN
cana-5267	3	2	a.	a.	PROPN
cana-5267	3	3	patil	patil	PROPN
cana-5267	3	4	1dept	1dept	NUM
cana-5267	3	5	of	of	ADP
cana-5267	3	6	computer	computer	NOUN
cana-5267	3	7	engineering	engineering	NOUN
cana-5267	3	8	,	,	PUNCT
cana-5267	3	9	jspm	jspm	PROPN
cana-5267	3	10	university	university	PROPN
cana-5267	3	11	pune	pune	NOUN
cana-5267	3	12	bgadade9000@gmail.com	bgadade9000@gmail.com	PUNCT
cana-5267	4	1	2dept	2dept	NUM
cana-5267	4	2	of	of	ADP
cana-5267	4	3	computer	computer	NOUN
cana-5267	4	4	engineering	engineering	NOUN
cana-5267	4	5	,	,	PUNCT
cana-5267	4	6	jspm	jspm	PROPN
cana-5267	4	7	university	university	PROPN
cana-5267	4	8	pune	pune	NOUN
cana-5267	4	9	gap.scos@jspmuni.ac.in	gap.scos@jspmuni.ac.in	PROPN
cana-5267	4	10	article	article	NOUN
cana-5267	4	11	history	history	NOUN
cana-5267	4	12	:	:	PUNCT
cana-5267	4	13	received	receive	VERB
cana-5267	4	14	:	:	PUNCT
cana-5267	4	15	12	12	NUM
cana-5267	4	16	-	-	SYM
cana-5267	4	17	01	01	NUM
cana-5267	4	18	-	-	PUNCT
cana-5267	4	19	2025	2025	NUM
cana-5267	4	20	revised	revise	VERB
cana-5267	4	21	:	:	PUNCT
cana-5267	4	22	15	15	NUM
cana-5267	4	23	-	-	NUM
cana-5267	4	24	02	02	NUM
cana-5267	4	25	-	-	PUNCT
cana-5267	4	26	2025	2025	NUM
cana-5267	4	27	accepted	accept	VERB
cana-5267	4	28	:	:	PUNCT
cana-5267	4	29	01	01	NUM
cana-5267	4	30	-	-	SYM
cana-5267	4	31	03	03	NUM
cana-5267	4	32	-	-	PUNCT
cana-5267	4	33	2025	2025	NUM
cana-5267	4	34	abstract	abstract	NOUN
cana-5267	4	35	:	:	PUNCT
cana-5267	4	36	with	with	ADP
cana-5267	4	37	the	the	DET
cana-5267	4	38	growing	grow	VERB
cana-5267	4	39	use	use	NOUN
cana-5267	4	40	of	of	ADP
cana-5267	4	41	large	large	ADJ
cana-5267	4	42	language	language	NOUN
cana-5267	4	43	models	model	NOUN
cana-5267	4	44	(	(	PUNCT
cana-5267	4	45	llms	llm	NOUN
cana-5267	4	46	)	)	PUNCT
cana-5267	4	47	in	in	ADP
cana-5267	4	48	diverse	diverse	ADJ
cana-5267	4	49	applications	application	NOUN
cana-5267	4	50	,	,	PUNCT
cana-5267	4	51	their	their	PRON
cana-5267	4	52	safety	safety	NOUN
cana-5267	4	53	,	,	PUNCT
cana-5267	4	54	security	security	NOUN
cana-5267	4	55	and	and	CCONJ
cana-5267	4	56	resilience	resilience	NOUN
cana-5267	4	57	against	against	ADP
cana-5267	4	58	cyber	cyber	NOUN
cana-5267	4	59	threats	threat	NOUN
cana-5267	4	60	has	have	AUX
cana-5267	4	61	become	become	VERB
cana-5267	4	62	increasingly	increasingly	ADV
cana-5267	4	63	worrying	worrying	ADJ
cana-5267	4	64	.	.	PUNCT
cana-5267	5	1	traditional	traditional	ADJ
cana-5267	5	2	security	security	NOUN
cana-5267	5	3	measures	measure	NOUN
cana-5267	5	4	routinely	routinely	ADV
cana-5267	5	5	lack	lack	VERB
cana-5267	5	6	to	to	PART
cana-5267	5	7	counter	counter	VERB
cana-5267	5	8	the	the	DET
cana-5267	5	9	active	active	ADJ
cana-5267	5	10	and	and	CCONJ
cana-5267	5	11	adaptive	adaptive	ADJ
cana-5267	5	12	nature	nature	NOUN
cana-5267	5	13	of	of	ADP
cana-5267	5	14	vulnerabilities	vulnerability	NOUN
cana-5267	5	15	within	within	ADP
cana-5267	5	16	llm	llm	PROPN
cana-5267	5	17	systems	system	NOUN
cana-5267	5	18	,	,	PUNCT
cana-5267	5	19	thus	thus	ADV
cana-5267	5	20	necessitating	necessitate	VERB
cana-5267	5	21	an	an	DET
cana-5267	5	22	automated	automate	VERB
cana-5267	5	23	threat	threat	NOUN
cana-5267	5	24	modeling	modeling	NOUN
cana-5267	5	25	(	(	PUNCT
cana-5267	5	26	atm	atm	NOUN
cana-5267	5	27	)	)	PUNCT
cana-5267	5	28	technique	technique	NOUN
cana-5267	5	29	.	.	PUNCT
cana-5267	6	1	this	this	DET
cana-5267	6	2	research	research	NOUN
cana-5267	6	3	introduces	introduce	VERB
cana-5267	6	4	a	a	DET
cana-5267	6	5	threat	threat	NOUN
cana-5267	6	6	modeling	model	VERB
cana-5267	6	7	framework	framework	NOUN
cana-5267	6	8	that	that	PRON
cana-5267	6	9	uses	use	VERB
cana-5267	6	10	ai	ai	AUX
cana-5267	6	11	tailored	tailor	VERB
cana-5267	6	12	for	for	ADP
cana-5267	6	13	complete	complete	ADJ
cana-5267	6	14	risk	risk	NOUN
cana-5267	6	15	management	management	NOUN
cana-5267	6	16	through	through	ADP
cana-5267	6	17	identification	identification	NOUN
cana-5267	6	18	,	,	PUNCT
cana-5267	6	19	evaluation	evaluation	NOUN
cana-5267	6	20	,	,	PUNCT
cana-5267	6	21	and	and	CCONJ
cana-5267	6	22	mitigation	mitigation	NOUN
cana-5267	6	23	of	of	ADP
cana-5267	6	24	security	security	NOUN
cana-5267	6	25	threats	threat	NOUN
cana-5267	6	26	stemming	stem	VERB
cana-5267	6	27	from	from	ADP
cana-5267	6	28	llm	llm	PROPN
cana-5267	6	29	system	system	NOUN
cana-5267	6	30	integrations	integration	NOUN
cana-5267	6	31	.	.	PUNCT
cana-5267	7	1	the	the	DET
cana-5267	7	2	proposed	propose	VERB
cana-5267	7	3	automated	automate	VERB
cana-5267	7	4	threat	threat	NOUN
cana-5267	7	5	modeling	modeling	NOUN
cana-5267	7	6	(	(	PUNCT
cana-5267	7	7	atm	atm	NOUN
cana-5267	7	8	)	)	PUNCT
cana-5267	7	9	system	system	NOUN
cana-5267	7	10	applies	apply	VERB
cana-5267	7	11	machine	machine	NOUN
cana-5267	7	12	learning	learning	NOUN
cana-5267	7	13	,	,	PUNCT
cana-5267	7	14	natural	natural	ADJ
cana-5267	7	15	language	language	NOUN
cana-5267	7	16	processing	processing	NOUN
cana-5267	7	17	(	(	PUNCT
cana-5267	7	18	nlp	nlp	NOUN
cana-5267	7	19	)	)	PUNCT
cana-5267	7	20	,	,	PUNCT
cana-5267	7	21	and	and	CCONJ
cana-5267	7	22	behavioural	behavioural	ADJ
cana-5267	7	23	scrutiny	scrutiny	NOUN
cana-5267	7	24	for	for	ADP
cana-5267	7	25	advanced	advanced	ADJ
cana-5267	7	26	diagnosis	diagnosis	NOUN
cana-5267	7	27	to	to	ADP
cana-5267	7	28	multi	multi	ADJ
cana-5267	7	29	-	-	ADJ
cana-5267	7	30	faceted	faceted	ADJ
cana-5267	7	31	emergent	emergent	ADJ
cana-5267	7	32	attack	attack	NOUN
cana-5267	7	33	spectrum	spectrum	NOUN
cana-5267	7	34	like	like	ADP
cana-5267	7	35	prompt	prompt	ADJ
cana-5267	7	36	injection	injection	NOUN
cana-5267	7	37	,	,	PUNCT
cana-5267	7	38	data	data	NOUN
cana-5267	7	39	poisoning	poisoning	NOUN
cana-5267	7	40	,	,	PUNCT
cana-5267	7	41	model	model	NOUN
cana-5267	7	42	inversion	inversion	NOUN
cana-5267	7	43	,	,	PUNCT
cana-5267	7	44	adversarial	adversarial	ADJ
cana-5267	7	45	attacks	attack	NOUN
cana-5267	7	46	,	,	PUNCT
cana-5267	7	47	and	and	CCONJ
cana-5267	7	48	unauthorized	unauthorized	ADJ
cana-5267	7	49	access	access	NOUN
cana-5267	7	50	.	.	PUNCT
cana-5267	8	1	a	a	DET
cana-5267	8	2	hybrid	hybrid	ADJ
cana-5267	8	3	risk	risk	NOUN
cana-5267	8	4	assessment	assessment	NOUN
cana-5267	8	5	approach	approach	NOUN
cana-5267	8	6	is	be	AUX
cana-5267	8	7	implanted	implant	VERB
cana-5267	8	8	in	in	ADP
cana-5267	8	9	the	the	DET
cana-5267	8	10	framework	framework	NOUN
cana-5267	8	11	,	,	PUNCT
cana-5267	8	12	using	use	VERB
cana-5267	8	13	static	static	ADJ
cana-5267	8	14	security	security	NOUN
cana-5267	8	15	assessment	assessment	NOUN
cana-5267	8	16	,	,	PUNCT
cana-5267	8	17	dynamic	dynamic	ADJ
cana-5267	8	18	behavioural	behavioural	ADJ
cana-5267	8	19	profiling	profiling	NOUN
cana-5267	8	20	,	,	PUNCT
cana-5267	8	21	as	as	ADV
cana-5267	8	22	well	well	ADV
cana-5267	8	23	as	as	ADP
cana-5267	8	24	real	real	ADJ
cana-5267	8	25	-	-	PUNCT
cana-5267	8	26	time	time	NOUN
cana-5267	8	27	anomaly	anomaly	NOUN
cana-5267	8	28	detection	detection	NOUN
cana-5267	8	29	to	to	PART
cana-5267	8	30	improve	improve	VERB
cana-5267	8	31	threat	threat	NOUN
cana-5267	8	32	detection	detection	NOUN
cana-5267	8	33	accuracy	accuracy	NOUN
cana-5267	8	34	.	.	PUNCT
cana-5267	9	1	moreover	moreover	ADV
cana-5267	9	2	,	,	PUNCT
cana-5267	9	3	the	the	DET
cana-5267	9	4	model	model	NOUN
cana-5267	9	5	adapts	adapt	VERB
cana-5267	9	6	cyber	cyber	NOUN
cana-5267	9	7	threat	threat	NOUN
cana-5267	9	8	intelligence	intelligence	NOUN
cana-5267	9	9	(	(	PUNCT
cana-5267	9	10	cti	cti	PROPN
cana-5267	9	11	)	)	PUNCT
cana-5267	9	12	feeds	feed	VERB
cana-5267	9	13	with	with	ADP
cana-5267	9	14	automated	automate	VERB
cana-5267	9	15	threat	threat	NOUN
cana-5267	9	16	neutralization	neutralization	NOUN
cana-5267	9	17	policies	policy	NOUN
cana-5267	9	18	for	for	ADP
cana-5267	9	19	advanced	advanced	ADJ
cana-5267	9	20	proactive	proactive	ADJ
cana-5267	9	21	defence	defence	NOUN
cana-5267	9	22	mechanisms	mechanism	NOUN
cana-5267	9	23	.	.	PUNCT
cana-5267	10	1	systematic	systematic	ADJ
cana-5267	10	2	testing	testing	NOUN
cana-5267	10	3	of	of	ADP
cana-5267	10	4	the	the	DET
cana-5267	10	5	model	model	NOUN
cana-5267	10	6	in	in	ADP
cana-5267	10	7	real	real	ADJ
cana-5267	10	8	-	-	PUNCT
cana-5267	10	9	world	world	NOUN
cana-5267	10	10	llm	llm	PROPN
cana-5267	10	11	deployment	deployment	NOUN
cana-5267	10	12	scenarios	scenario	NOUN
cana-5267	10	13	validated	validate	VERB
cana-5267	10	14	its	its	PRON
cana-5267	10	15	efficacy	efficacy	NOUN
cana-5267	10	16	in	in	ADP
cana-5267	10	17	precision	precision	NOUN
cana-5267	10	18	and	and	CCONJ
cana-5267	10	19	recall	recall	NOUN
cana-5267	10	20	metrics	metric	NOUN
cana-5267	10	21	for	for	ADP
cana-5267	10	22	securing	secure	VERB
cana-5267	10	23	vulnerability	vulnerability	NOUN
cana-5267	10	24	detection	detection	NOUN
cana-5267	10	25	and	and	CCONJ
cana-5267	10	26	mitigation	mitigation	NOUN
cana-5267	10	27	.	.	PUNCT
cana-5267	11	1	findings	finding	NOUN
cana-5267	11	2	demonstrate	demonstrate	VERB
cana-5267	11	3	the	the	DET
cana-5267	11	4	efficacy	efficacy	NOUN
cana-5267	11	5	of	of	ADP
cana-5267	11	6	ai	ai	PROPN
cana-5267	11	7	driven	drive	VERB
cana-5267	11	8	threat	threat	NOUN
cana-5267	11	9	modeling	modeling	NOUN
cana-5267	11	10	in	in	ADP
cana-5267	11	11	llm	llm	PROPN
cana-5267	11	12	integrated	integrate	VERB
cana-5267	11	13	systems	system	NOUN
cana-5267	11	14	for	for	ADP
cana-5267	11	15	automation	automation	NOUN
cana-5267	11	16	of	of	ADP
cana-5267	11	17	risk	risk	NOUN
cana-5267	11	18	assessment	assessment	NOUN
cana-5267	11	19	,	,	PUNCT
cana-5267	11	20	decreased	decrease	VERB
cana-5267	11	21	false	false	ADJ
cana-5267	11	22	positives	positive	NOUN
cana-5267	11	23	,	,	PUNCT
cana-5267	11	24	and	and	CCONJ
cana-5267	11	25	improved	improved	ADJ
cana-5267	11	26	response	response	NOUN
cana-5267	11	27	time	time	NOUN
cana-5267	11	28	while	while	SCONJ
cana-5267	11	29	fortifying	fortifying	ADJ
cana-5267	11	30	system	system	NOUN
cana-5267	11	31	security	security	NOUN
cana-5267	11	32	.	.	PUNCT
cana-5267	12	1	this	this	DET
cana-5267	12	2	study	study	NOUN
cana-5267	12	3	highlights	highlight	VERB
cana-5267	12	4	the	the	DET
cana-5267	12	5	need	need	NOUN
cana-5267	12	6	for	for	ADP
cana-5267	12	7	implementing	implement	VERB
cana-5267	12	8	proactive	proactive	ADJ
cana-5267	12	9	and	and	CCONJ
cana-5267	12	10	adaptive	adaptive	ADJ
cana-5267	12	11	security	security	NOUN
cana-5267	12	12	measures	measure	NOUN
cana-5267	12	13	during	during	ADP
cana-5267	12	14	the	the	DET
cana-5267	12	15	integration	integration	NOUN
cana-5267	12	16	of	of	ADP
cana-5267	12	17	llm	llm	PROPN
cana-5267	12	18	systems	system	NOUN
cana-5267	12	19	to	to	PART
cana-5267	12	20	maintain	maintain	VERB
cana-5267	12	21	the	the	DET
cana-5267	12	22	integrity	integrity	NOUN
cana-5267	12	23	and	and	CCONJ
cana-5267	12	24	reliability	reliability	NOUN
cana-5267	12	25	of	of	ADP
cana-5267	12	26	ai	ai	ADJ
cana-5267	12	27	applications	application	NOUN
cana-5267	12	28	amid	amid	ADP
cana-5267	12	29	shifting	shift	VERB
cana-5267	12	30	and	and	CCONJ
cana-5267	12	31	escalating	escalate	VERB
cana-5267	12	32	cyber	cyber	NOUN
cana-5267	12	33	risks	risk	NOUN
cana-5267	12	34	.	.	PUNCT
cana-5267	13	1	the	the	DET
cana-5267	13	2	advanced	advanced	ADJ
cana-5267	13	3	development	development	NOUN
cana-5267	13	4	in	in	ADP
cana-5267	13	5	secure	secure	ADJ
cana-5267	13	6	ai	ai	VERB
cana-5267	13	7	implementation	implementation	NOUN
cana-5267	13	8	strategies	strategy	NOUN
cana-5267	13	9	is	be	AUX
cana-5267	13	10	achieved	achieve	VERB
cana-5267	13	11	with	with	ADP
cana-5267	13	12	the	the	DET
cana-5267	13	13	automated	automate	VERB
cana-5267	13	14	threat	threat	NOUN
cana-5267	13	15	modeling	model	VERB
cana-5267	13	16	framework	framework	NOUN
cana-5267	13	17	supporting	support	VERB
cana-5267	13	18	the	the	DET
cana-5267	13	19	development	development	NOUN
cana-5267	13	20	of	of	ADP
cana-5267	13	21	future	future	ADJ
cana-5267	13	22	ai	ai	NOUN
cana-5267	13	23	-	-	PUNCT
cana-5267	13	24	based	base	VERB
cana-5267	13	25	cybersecurity	cybersecurity	NOUN
cana-5267	13	26	technologies	technology	NOUN
cana-5267	13	27	for	for	ADP
cana-5267	13	28	enterprise	enterprise	NOUN
cana-5267	13	29	and	and	CCONJ
cana-5267	13	30	cloud	cloud	ADJ
cana-5267	13	31	llm	llm	PROPN
cana-5267	13	32	applications	application	NOUN
cana-5267	13	33	.	.	PUNCT
cana-5267	14	1	keywords	keyword	NOUN
cana-5267	14	2	:	:	PUNCT
cana-5267	14	3	automated	automate	VERB
cana-5267	14	4	threat	threat	NOUN
cana-5267	14	5	modeling	modeling	NOUN
cana-5267	14	6	,	,	PUNCT
cana-5267	14	7	large	large	ADJ
cana-5267	14	8	language	language	NOUN
cana-5267	14	9	models	model	NOUN
cana-5267	14	10	,	,	PUNCT
cana-5267	14	11	cybersecurity	cybersecurity	NOUN
cana-5267	14	12	,	,	PUNCT
cana-5267	14	13	ai	ai	VERB
cana-5267	14	14	security	security	NOUN
cana-5267	14	15	,	,	PUNCT
cana-5267	14	16	machine	machine	NOUN
cana-5267	14	17	learning	learning	NOUN
cana-5267	14	18	,	,	PUNCT
cana-5267	14	19	risk	risk	NOUN
cana-5267	14	20	assessment	assessment	NOUN
cana-5267	14	21	,	,	PUNCT
cana-5267	14	22	adversarial	adversarial	ADJ
cana-5267	14	23	attacks	attack	NOUN
cana-5267	14	24	,	,	PUNCT
cana-5267	14	25	threat	threat	NOUN
cana-5267	14	26	intelligence	intelligence	NOUN
cana-5267	14	27	.	.	PUNCT
cana-5267	15	1	introduction	introduction	NOUN
cana-5267	15	2	the	the	DET
cana-5267	15	3	adoption	adoption	NOUN
cana-5267	15	4	of	of	ADP
cana-5267	15	5	large	large	ADJ
cana-5267	15	6	language	language	NOUN
cana-5267	15	7	models	model	NOUN
cana-5267	15	8	(	(	PUNCT
cana-5267	15	9	llms	llm	NOUN
cana-5267	15	10	)	)	PUNCT
cana-5267	15	11	like	like	VERB
cana-5267	15	12	open	open	ADJ
cana-5267	15	13	ai	ai	PROPN
cana-5267	15	14	’s	’s	PART
cana-5267	15	15	gpt	gpt	NOUN
cana-5267	15	16	series	series	NOUN
cana-5267	15	17	,	,	PUNCT
cana-5267	15	18	google	google	PROPN
cana-5267	15	19	’s	’s	PART
cana-5267	15	20	bard	bard	NOUN
cana-5267	15	21	,	,	PUNCT
cana-5267	15	22	and	and	CCONJ
cana-5267	15	23	even	even	ADV
cana-5267	15	24	meta	meta	X
cana-5267	15	25	’s	’s	ADJ
cana-5267	15	26	llama	llama	NOUN
cana-5267	15	27	,	,	PUNCT
cana-5267	15	28	have	have	VERB
cana-5267	15	29	streamlined	streamline	VERB
cana-5267	15	30	industries	industry	NOUN
cana-5267	15	31	such	such	ADJ
cana-5267	15	32	as	as	ADP
cana-5267	15	33	finance	finance	NOUN
cana-5267	15	34	,	,	PUNCT
cana-5267	15	35	healthcare	healthcare	PROPN
cana-5267	15	36	,	,	PUNCT
cana-5267	15	37	cybersecurity	cybersecurity	NOUN
cana-5267	15	38	,	,	PUNCT
cana-5267	15	39	and	and	CCONJ
cana-5267	15	40	even	even	ADV
cana-5267	15	41	automated	automate	VERB
cana-5267	15	42	decision	decision	NOUN
cana-5267	15	43	making	making	NOUN
cana-5267	15	44	.	.	PUNCT
cana-5267	16	1	their	their	PRON
cana-5267	16	2	astonishing	astonishing	ADJ
cana-5267	16	3	capabilities	capability	NOUN
cana-5267	16	4	in	in	ADP
cana-5267	16	5	natural	natural	ADJ
cana-5267	16	6	language	language	NOUN
cana-5267	16	7	processing	processing	NOUN
cana-5267	16	8	,	,	PUNCT
cana-5267	16	9	mailto:bgadade9000@gmail.com	mailto:bgadade9000@gmail.com	X
cana-5267	16	10	mailto:gap.scos@jspmuni.ac.in	mailto:gap.scos@jspmuni.ac.in	PROPN
cana-5267	16	11	communications	communication	NOUN
cana-5267	16	12	on	on	ADP
cana-5267	16	13	applied	apply	VERB
cana-5267	16	14	nonlinear	nonlinear	ADJ
cana-5267	16	15	analysis	analysis	NOUN
cana-5267	16	16	issn	issn	NOUN
cana-5267	16	17	:	:	PUNCT
cana-5267	16	18	1074	1074	NUM
cana-5267	16	19	-	-	PUNCT
cana-5267	16	20	133x	133x	NUM
cana-5267	16	21	vol	vol	VERB
cana-5267	16	22	32	32	NUM
cana-5267	16	23	no	no	NOUN
cana-5267	16	24	.	.	PUNCT
cana-5267	17	1	10s	10	NOUN
cana-5267	17	2	(	(	PUNCT
cana-5267	17	3	2025	2025	NUM
cana-5267	17	4	)	)	PUNCT
cana-5267	17	5	1659	1659	NUM
cana-5267	17	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5267	17	7	generation	generation	NOUN
cana-5267	17	8	,	,	PUNCT
cana-5267	17	9	and	and	CCONJ
cana-5267	17	10	reasoning	reasoning	NOUN
cana-5267	17	11	helps	help	VERB
cana-5267	17	12	to	to	PART
cana-5267	17	13	accomplish	accomplish	VERB
cana-5267	17	14	this	this	PRON
cana-5267	17	15	.	.	PUNCT
cana-5267	18	1	however	however	ADV
cana-5267	18	2	,	,	PUNCT
cana-5267	18	3	due	due	ADP
cana-5267	18	4	to	to	ADP
cana-5267	18	5	the	the	DET
cana-5267	18	6	presence	presence	NOUN
cana-5267	18	7	of	of	ADP
cana-5267	18	8	complex	complex	ADJ
cana-5267	18	9	security	security	NOUN
cana-5267	18	10	risks	risk	NOUN
cana-5267	18	11	like	like	ADP
cana-5267	18	12	adversarial	adversarial	ADJ
cana-5267	18	13	attacks	attack	NOUN
cana-5267	18	14	,	,	PUNCT
cana-5267	18	15	data	datum	NOUN
cana-5267	18	16	leakage	leakage	NOUN
cana-5267	18	17	,	,	PUNCT
cana-5267	18	18	bias	bias	NOUN
cana-5267	18	19	exploitation	exploitation	NOUN
cana-5267	18	20	,	,	PUNCT
cana-5267	18	21	and	and	CCONJ
cana-5267	18	22	prompt	prompt	ADJ
cana-5267	18	23	injection	injection	NOUN
cana-5267	18	24	vulnerabilities	vulnerability	NOUN
cana-5267	18	25	,	,	PUNCT
cana-5267	18	26	the	the	DET
cana-5267	18	27	integration	integration	NOUN
cana-5267	18	28	of	of	ADP
cana-5267	18	29	these	these	DET
cana-5267	18	30	models	model	NOUN
cana-5267	18	31	into	into	ADP
cana-5267	18	32	more	more	ADV
cana-5267	18	33	sophisticated	sophisticated	ADJ
cana-5267	18	34	systems	system	NOUN
cana-5267	18	35	can	can	AUX
cana-5267	18	36	be	be	AUX
cana-5267	18	37	problematic	problematic	ADJ
cana-5267	18	38	.	.	PUNCT
cana-5267	19	1	reliance	reliance	NOUN
cana-5267	19	2	on	on	ADP
cana-5267	19	3	llms	llm	NOUN
cana-5267	19	4	to	to	PART
cana-5267	19	5	perform	perform	VERB
cana-5267	19	6	important	important	ADJ
cana-5267	19	7	mission	mission	NOUN
cana-5267	19	8	-	-	PUNCT
cana-5267	19	9	critical	critical	ADJ
cana-5267	19	10	tasks	task	NOUN
cana-5267	19	11	increases	increase	VERB
cana-5267	19	12	the	the	DET
cana-5267	19	13	risk	risk	NOUN
cana-5267	19	14	of	of	ADP
cana-5267	19	15	security	security	NOUN
cana-5267	19	16	threats	threat	NOUN
cana-5267	19	17	.	.	PUNCT
cana-5267	20	1	this	this	PRON
cana-5267	20	2	compromises	compromise	VERB
cana-5267	20	3	the	the	DET
cana-5267	20	4	system	system	NOUN
cana-5267	20	5	’s	’s	PART
cana-5267	20	6	protective	protective	ADJ
cana-5267	20	7	layers	layer	NOUN
cana-5267	20	8	guaranteeing	guarantee	VERB
cana-5267	20	9	control	control	NOUN
cana-5267	20	10	,	,	PUNCT
cana-5267	20	11	secrecy	secrecy	NOUN
cana-5267	20	12	,	,	PUNCT
cana-5267	20	13	and	and	CCONJ
cana-5267	20	14	availability	availability	NOUN
cana-5267	20	15	,	,	PUNCT
cana-5267	20	16	calling	call	VERB
cana-5267	20	17	for	for	ADP
cana-5267	20	18	stricter	strict	ADJ
cana-5267	20	19	policies	policy	NOUN
cana-5267	20	20	.	.	PUNCT
cana-5267	21	1	automated	automate	VERB
cana-5267	21	2	threat	threat	NOUN
cana-5267	21	3	modeling	modeling	NOUN
cana-5267	21	4	(	(	PUNCT
cana-5267	21	5	atm	atm	NOUN
cana-5267	21	6	)	)	PUNCT
cana-5267	21	7	tackles	tackle	VERB
cana-5267	21	8	llm	llm	PROPN
cana-5267	21	9	risks	risk	NOUN
cana-5267	21	10	head	head	NOUN
cana-5267	21	11	on	on	ADP
cana-5267	21	12	by	by	ADP
cana-5267	21	13	identifying	identify	VERB
cana-5267	21	14	and	and	CCONJ
cana-5267	21	15	assessing	assess	VERB
cana-5267	21	16	these	these	DET
cana-5267	21	17	problems	problem	NOUN
cana-5267	21	18	in	in	ADP
cana-5267	21	19	advance	advance	NOUN
cana-5267	21	20	.	.	PUNCT
cana-5267	22	1	using	use	VERB
cana-5267	22	2	ai	ai	NOUN
cana-5267	22	3	-	-	PUNCT
cana-5267	22	4	based	base	VERB
cana-5267	22	5	threat	threat	NOUN
cana-5267	22	6	modeling	model	VERB
cana-5267	22	7	frameworks	framework	NOUN
cana-5267	22	8	equips	equip	VERB
cana-5267	22	9	organizations	organization	NOUN
cana-5267	22	10	with	with	ADP
cana-5267	22	11	foresight	foresight	NOUN
cana-5267	22	12	through	through	ADP
cana-5267	22	13	potential	potential	ADJ
cana-5267	22	14	vulnerability	vulnerability	NOUN
cana-5267	22	15	assessment	assessment	NOUN
cana-5267	22	16	enabled	enable	VERB
cana-5267	22	17	by	by	ADP
cana-5267	22	18	strategically	strategically	ADV
cana-5267	22	19	planned	plan	VERB
cana-5267	22	20	defenses	defense	NOUN
cana-5267	22	21	before	before	ADP
cana-5267	22	22	facing	face	VERB
cana-5267	22	23	active	active	ADJ
cana-5267	22	24	threats	threat	NOUN
cana-5267	22	25	from	from	ADP
cana-5267	22	26	rivals	rival	NOUN
cana-5267	22	27	.	.	PUNCT
cana-5267	23	1	microsoft	microsoft	PROPN
cana-5267	23	2	's	's	PART
cana-5267	23	3	stride	stride	NOUN
cana-5267	23	4	,	,	PUNCT
cana-5267	23	5	mitre	mitre	NOUN
cana-5267	23	6	att&ck	att&ck	NOUN
cana-5267	23	7	,	,	PUNCT
cana-5267	23	8	and	and	CCONJ
cana-5267	23	9	nist	nist	NOUN
cana-5267	23	10	cybersecurity	cybersecurity	NOUN
cana-5267	23	11	frameworks	framework	NOUN
cana-5267	23	12	focus	focus	VERB
cana-5267	23	13	on	on	ADP
cana-5267	23	14	systematic	systematic	ADJ
cana-5267	23	15	approaches	approach	NOUN
cana-5267	23	16	to	to	PART
cana-5267	23	17	cyber	cyber	VERB
cana-5267	23	18	security	security	NOUN
cana-5267	23	19	and	and	CCONJ
cana-5267	23	20	software	software	NOUN
cana-5267	23	21	security	security	NOUN
cana-5267	23	22	diagnostics	diagnostic	NOUN
cana-5267	23	23	.	.	PUNCT
cana-5267	24	1	these	these	DET
cana-5267	24	2	techniques	technique	NOUN
cana-5267	24	3	have	have	AUX
cana-5267	24	4	been	be	AUX
cana-5267	24	5	integrated	integrate	VERB
cana-5267	24	6	into	into	ADP
cana-5267	24	7	software	software	NOUN
cana-5267	24	8	security	security	NOUN
cana-5267	24	9	analysis	analysis	NOUN
cana-5267	24	10	.	.	PUNCT
cana-5267	25	1	however	however	ADV
cana-5267	25	2	,	,	PUNCT
cana-5267	25	3	the	the	DET
cana-5267	25	4	structure	structure	NOUN
cana-5267	25	5	and	and	CCONJ
cana-5267	25	6	functioning	functioning	NOUN
cana-5267	25	7	of	of	ADP
cana-5267	25	8	llms	llm	NOUN
cana-5267	25	9	bring	bring	VERB
cana-5267	25	10	about	about	ADV
cana-5267	25	11	additional	additional	ADJ
cana-5267	25	12	attack	attack	NOUN
cana-5267	25	13	vectors	vector	NOUN
cana-5267	25	14	that	that	PRON
cana-5267	25	15	require	require	VERB
cana-5267	25	16	distinct	distinct	ADJ
cana-5267	25	17	threat	threat	NOUN
cana-5267	25	18	modeling	model	VERB
cana-5267	25	19	approaches	approach	NOUN
cana-5267	25	20	.	.	PUNCT
cana-5267	26	1	different	different	ADJ
cana-5267	26	2	from	from	ADP
cana-5267	26	3	software	software	NOUN
cana-5267	26	4	parts	part	NOUN
cana-5267	26	5	,	,	PUNCT
cana-5267	26	6	llms	llm	NOUN
cana-5267	26	7	respond	respond	VERB
cana-5267	26	8	actively	actively	ADV
cana-5267	26	9	to	to	ADP
cana-5267	26	10	databases	database	NOUN
cana-5267	26	11	,	,	PUNCT
cana-5267	26	12	external	external	ADJ
cana-5267	26	13	apis	apis	NOUN
cana-5267	26	14	,	,	PUNCT
cana-5267	26	15	and	and	CCONJ
cana-5267	26	16	user	user	NOUN
cana-5267	26	17	inputs	input	NOUN
cana-5267	26	18	,	,	PUNCT
cana-5267	26	19	which	which	PRON
cana-5267	26	20	makes	make	VERB
cana-5267	26	21	them	they	PRON
cana-5267	26	22	susceptible	susceptible	ADJ
cana-5267	26	23	to	to	ADP
cana-5267	26	24	prompt	prompt	VERB
cana-5267	26	25	engineering	engineering	NOUN
cana-5267	26	26	attacks	attack	NOUN
cana-5267	26	27	,	,	PUNCT
cana-5267	26	28	adversarial	adversarial	ADJ
cana-5267	26	29	perturbation	perturbation	NOUN
cana-5267	26	30	,	,	PUNCT
cana-5267	26	31	data	data	NOUN
cana-5267	26	32	exfiltration	exfiltration	NOUN
cana-5267	26	33	,	,	PUNCT
cana-5267	26	34	and	and	CCONJ
cana-5267	26	35	model	model	NOUN
cana-5267	26	36	manipulation	manipulation	NOUN
cana-5267	26	37	.	.	PUNCT
cana-5267	27	1	it	it	PRON
cana-5267	27	2	is	be	AUX
cana-5267	27	3	crucial	crucial	ADJ
cana-5267	27	4	to	to	PART
cana-5267	27	5	incorporate	incorporate	VERB
cana-5267	27	6	automated	automate	VERB
cana-5267	27	7	threat	threat	NOUN
cana-5267	27	8	modeling	modeling	NOUN
cana-5267	27	9	(	(	PUNCT
cana-5267	27	10	atm	atm	NOUN
cana-5267	27	11	)	)	PUNCT
cana-5267	27	12	during	during	ADP
cana-5267	27	13	llm	llm	PROPN
cana-5267	27	14	system	system	NOUN
cana-5267	27	15	deployment	deployment	NOUN
cana-5267	27	16	to	to	PART
cana-5267	27	17	cover	cover	VERB
cana-5267	27	18	security	security	NOUN
cana-5267	27	19	loopholes	loophole	NOUN
cana-5267	27	20	,	,	PUNCT
cana-5267	27	21	strengthen	strengthen	VERB
cana-5267	27	22	system	system	NOUN
cana-5267	27	23	defense	defense	NOUN
cana-5267	27	24	,	,	PUNCT
cana-5267	27	25	and	and	CCONJ
cana-5267	27	26	bolster	bolster	VERB
cana-5267	27	27	ai	ai	VERB
cana-5267	27	28	infrastructure	infrastructure	NOUN
cana-5267	27	29	against	against	ADP
cana-5267	27	30	advanced	advanced	ADJ
cana-5267	27	31	cyber	cyber	NOUN
cana-5267	27	32	attacks.the	attacks.the	DET
cana-5267	27	33	incorporation	incorporation	NOUN
cana-5267	27	34	of	of	ADP
cana-5267	27	35	llms	llm	NOUN
cana-5267	27	36	in	in	ADP
cana-5267	27	37	enterprise	enterprise	NOUN
cana-5267	27	38	applications	application	NOUN
cana-5267	27	39	,	,	PUNCT
cana-5267	27	40	chatbots	chatbot	NOUN
cana-5267	27	41	,	,	PUNCT
cana-5267	27	42	financial	financial	ADJ
cana-5267	27	43	automation	automation	NOUN
cana-5267	27	44	,	,	PUNCT
cana-5267	27	45	and	and	CCONJ
cana-5267	27	46	ever	ever	ADV
cana-5267	27	47	more	more	ADV
cana-5267	27	48	sophisticated	sophisticated	ADJ
cana-5267	27	49	cloud	cloud	NOUN
cana-5267	27	50	systems	system	NOUN
cana-5267	27	51	has	have	AUX
cana-5267	27	52	broadened	broaden	VERB
cana-5267	27	53	their	their	PRON
cana-5267	27	54	scope	scope	NOUN
cana-5267	27	55	of	of	ADP
cana-5267	27	56	use	use	NOUN
cana-5267	27	57	.	.	PUNCT
cana-5267	28	1	unfortunately	unfortunately	ADV
cana-5267	28	2	,	,	PUNCT
cana-5267	28	3	it	it	PRON
cana-5267	28	4	has	have	AUX
cana-5267	28	5	also	also	ADV
cana-5267	28	6	made	make	VERB
cana-5267	28	7	them	they	PRON
cana-5267	28	8	more	more	ADV
cana-5267	28	9	vulnerable	vulnerable	ADJ
cana-5267	28	10	to	to	ADP
cana-5267	28	11	a	a	DET
cana-5267	28	12	diverse	diverse	ADJ
cana-5267	28	13	set	set	NOUN
cana-5267	28	14	of	of	ADP
cana-5267	28	15	security	security	NOUN
cana-5267	28	16	challenges	challenge	NOUN
cana-5267	28	17	.	.	PUNCT
cana-5267	29	1	one	one	NUM
cana-5267	29	2	of	of	ADP
cana-5267	29	3	the	the	DET
cana-5267	29	4	major	major	ADJ
cana-5267	29	5	issues	issue	NOUN
cana-5267	29	6	is	be	AUX
cana-5267	29	7	prompt	prompt	ADJ
cana-5267	29	8	injection	injection	NOUN
cana-5267	29	9	attacks	attack	NOUN
cana-5267	29	10	,	,	PUNCT
cana-5267	29	11	where	where	SCONJ
cana-5267	29	12	attackers	attacker	NOUN
cana-5267	29	13	design	design	VERB
cana-5267	29	14	manipulative	manipulative	ADJ
cana-5267	29	15	input	input	NOUN
cana-5267	29	16	prompts	prompt	NOUN
cana-5267	29	17	with	with	ADP
cana-5267	29	18	the	the	DET
cana-5267	29	19	intent	intent	NOUN
cana-5267	29	20	of	of	ADP
cana-5267	29	21	triggering	trigger	VERB
cana-5267	29	22	covert	covert	ADJ
cana-5267	29	23	actions	action	NOUN
cana-5267	29	24	that	that	PRON
cana-5267	29	25	lead	lead	VERB
cana-5267	29	26	to	to	ADP
cana-5267	29	27	freely	freely	ADV
cana-5267	29	28	sharing	share	VERB
cana-5267	29	29	sensitive	sensitive	ADJ
cana-5267	29	30	information	information	NOUN
cana-5267	29	31	,	,	PUNCT
cana-5267	29	32	executing	execute	VERB
cana-5267	29	33	harmful	harmful	ADJ
cana-5267	29	34	instructions	instruction	NOUN
cana-5267	29	35	,	,	PUNCT
cana-5267	29	36	or	or	CCONJ
cana-5267	29	37	providing	provide	VERB
cana-5267	29	38	undesired	undesired	ADJ
cana-5267	29	39	answers	answer	NOUN
cana-5267	29	40	.	.	PUNCT
cana-5267	30	1	another	another	DET
cana-5267	30	2	important	important	ADJ
cana-5267	30	3	model	model	NOUN
cana-5267	30	4	inversion	inversion	NOUN
cana-5267	30	5	attacks	attack	NOUN
cana-5267	30	6	pose	pose	VERB
cana-5267	30	7	serious	serious	ADJ
cana-5267	30	8	risk	risk	NOUN
cana-5267	30	9	whereby	whereby	SCONJ
cana-5267	30	10	adversaries	adversary	NOUN
cana-5267	30	11	iteratively	iteratively	ADV
cana-5267	30	12	query	query	VERB
cana-5267	30	13	a	a	DET
cana-5267	30	14	model	model	NOUN
cana-5267	30	15	to	to	PART
cana-5267	30	16	reconstruct	reconstruct	VERB
cana-5267	30	17	the	the	DET
cana-5267	30	18	training	training	NOUN
cana-5267	30	19	data	datum	NOUN
cana-5267	30	20	.	.	PUNCT
cana-5267	31	1	this	this	PRON
cana-5267	31	2	may	may	AUX
cana-5267	31	3	lead	lead	VERB
cana-5267	31	4	to	to	ADP
cana-5267	31	5	privacy	privacy	NOUN
cana-5267	31	6	violations	violation	NOUN
cana-5267	31	7	and	and	CCONJ
cana-5267	31	8	unauthorized	unauthorized	ADJ
cana-5267	31	9	data	datum	NOUN
cana-5267	31	10	exposure	exposure	NOUN
cana-5267	31	11	.	.	PUNCT
cana-5267	32	1	moreover	moreover	ADV
cana-5267	32	2	,	,	PUNCT
cana-5267	32	3	adversarial	adversarial	ADJ
cana-5267	32	4	attacks	attack	NOUN
cana-5267	32	5	that	that	PRON
cana-5267	32	6	rely	rely	VERB
cana-5267	32	7	on	on	ADP
cana-5267	32	8	specific	specific	ADJ
cana-5267	32	9	modifications	modification	NOUN
cana-5267	32	10	designed	design	VERB
cana-5267	32	11	to	to	PART
cana-5267	32	12	cause	cause	VERB
cana-5267	32	13	misleading	misleading	ADJ
cana-5267	32	14	outputs	output	NOUN
cana-5267	32	15	can	can	AUX
cana-5267	32	16	result	result	VERB
cana-5267	32	17	in	in	ADP
cana-5267	32	18	llms	llm	NOUN
cana-5267	32	19	making	make	VERB
cana-5267	32	20	errors	error	NOUN
cana-5267	32	21	that	that	PRON
cana-5267	32	22	can	can	AUX
cana-5267	32	23	be	be	AUX
cana-5267	32	24	catastrophic	catastrophic	ADJ
cana-5267	32	25	in	in	ADP
cana-5267	32	26	crucial	crucial	ADJ
cana-5267	32	27	areas	area	NOUN
cana-5267	32	28	like	like	ADP
cana-5267	32	29	autonomous	autonomous	ADJ
cana-5267	32	30	systems	system	NOUN
cana-5267	32	31	,	,	PUNCT
cana-5267	32	32	legal	legal	ADJ
cana-5267	32	33	consultancy	consultancy	NOUN
cana-5267	32	34	,	,	PUNCT
cana-5267	32	35	or	or	CCONJ
cana-5267	32	36	medical	medical	ADJ
cana-5267	32	37	diagnostics	diagnostic	NOUN
cana-5267	32	38	.	.	PUNCT
cana-5267	33	1	addressing	address	VERB
cana-5267	33	2	these	these	DET
cana-5267	33	3	gaps	gap	NOUN
cana-5267	33	4	entails	entail	NOUN
cana-5267	33	5	using	use	VERB
cana-5267	33	6	automated	automate	VERB
cana-5267	33	7	procedures	procedure	NOUN
cana-5267	33	8	for	for	ADP
cana-5267	33	9	constructing	construct	VERB
cana-5267	33	10	threat	threat	NOUN
cana-5267	33	11	models	model	NOUN
cana-5267	33	12	as	as	ADV
cana-5267	33	13	well	well	ADV
cana-5267	33	14	as	as	ADP
cana-5267	33	15	providing	provide	VERB
cana-5267	33	16	methods	method	NOUN
cana-5267	33	17	which	which	PRON
cana-5267	33	18	help	help	VERB
cana-5267	33	19	professionals	professional	NOUN
cana-5267	33	20	avert	avert	ADJ
cana-5267	33	21	,	,	PUNCT
cana-5267	33	22	identify	identify	VERB
cana-5267	33	23	,	,	PUNCT
cana-5267	33	24	and	and	CCONJ
cana-5267	33	25	neutralize	neutralize	VERB
cana-5267	33	26	possible	possible	ADJ
cana-5267	33	27	weaknesses	weakness	NOUN
cana-5267	33	28	advance	advance	NOUN
cana-5267	33	29	.	.	PUNCT
cana-5267	34	1	the	the	DET
cana-5267	34	2	combination	combination	NOUN
cana-5267	34	3	of	of	ADP
cana-5267	34	4	an	an	DET
cana-5267	34	5	llm	llm	NOUN
cana-5267	34	6	system	system	NOUN
cana-5267	34	7	with	with	ADP
cana-5267	34	8	machine	machine	NOUN
cana-5267	34	9	learning	learning	NOUN
cana-5267	34	10	,	,	PUNCT
cana-5267	34	11	ai	ai	VERB
cana-5267	34	12	,	,	PUNCT
cana-5267	34	13	cybersecurity	cybersecurity	NOUN
cana-5267	34	14	technologies	technology	NOUN
cana-5267	34	15	,	,	PUNCT
cana-5267	34	16	and	and	CCONJ
cana-5267	34	17	automated	automate	VERB
cana-5267	34	18	risk	risk	NOUN
cana-5267	34	19	evaluation	evaluation	NOUN
cana-5267	34	20	approaches	approach	NOUN
cana-5267	34	21	is	be	AUX
cana-5267	34	22	termed	term	VERB
cana-5267	34	23	automated	automate	VERB
cana-5267	34	24	threat	threat	NOUN
cana-5267	34	25	modeling	modeling	NOUN
cana-5267	34	26	(	(	PUNCT
cana-5267	34	27	atm	atm	NOUN
cana-5267	34	28	)	)	PUNCT
cana-5267	34	29	,	,	PUNCT
cana-5267	34	30	and	and	CCONJ
cana-5267	34	31	it	it	PRON
cana-5267	34	32	employs	employ	VERB
cana-5267	34	33	these	these	DET
cana-5267	34	34	integration	integration	NOUN
cana-5267	34	35	for	for	ADP
cana-5267	34	36	cyber	cyber	PROPN
cana-5267	34	37	proactive	proactive	ADJ
cana-5267	34	38	security	security	NOUN
cana-5267	34	39	.	.	PUNCT
cana-5267	35	1	in	in	ADP
cana-5267	35	2	contrast	contrast	NOUN
cana-5267	35	3	to	to	ADP
cana-5267	35	4	manual	manual	ADJ
cana-5267	35	5	atm	atm	NOUN
cana-5267	35	6	,	,	PUNCT
cana-5267	35	7	which	which	PRON
cana-5267	35	8	relies	rely	VERB
cana-5267	35	9	on	on	ADP
cana-5267	35	10	heuristic	heuristic	ADJ
cana-5267	35	11	reasoning	reasoning	NOUN
cana-5267	35	12	,	,	PUNCT
cana-5267	35	13	threat	threat	NOUN
cana-5267	35	14	modeling	modeling	NOUN
cana-5267	35	15	for	for	ADP
cana-5267	35	16	atm	atm	NOUN
cana-5267	35	17	relies	rely	VERB
cana-5267	35	18	on	on	ADP
cana-5267	35	19	automated	automate	VERB
cana-5267	35	20	frameworks	framework	NOUN
cana-5267	35	21	,	,	PUNCT
cana-5267	35	22	risk	risk	NOUN
cana-5267	35	23	analysis	analysis	NOUN
cana-5267	35	24	,	,	PUNCT
cana-5267	35	25	and	and	CCONJ
cana-5267	35	26	advanced	advanced	ADJ
cana-5267	35	27	dynamic	dynamic	ADJ
cana-5267	35	28	vulnerability	vulnerability	NOUN
cana-5267	35	29	scanning	scan	VERB
cana-5267	35	30	techniques	technique	NOUN
cana-5267	35	31	for	for	ADP
cana-5267	35	32	expeditious	expeditious	ADJ
cana-5267	35	33	security	security	NOUN
cana-5267	35	34	evaluation	evaluation	NOUN
cana-5267	35	35	.	.	PUNCT
cana-5267	36	1	atm	atm	PROPN
cana-5267	36	2	applies	apply	VERB
cana-5267	36	3	graph	graph	NOUN
cana-5267	36	4	-	-	PUNCT
cana-5267	36	5	based	base	VERB
cana-5267	36	6	security	security	NOUN
cana-5267	36	7	analysis	analysis	NOUN
cana-5267	36	8	,	,	PUNCT
cana-5267	36	9	behavior	behavior	NOUN
cana-5267	36	10	anomaly	anomaly	NOUN
cana-5267	36	11	detection	detection	NOUN
cana-5267	36	12	,	,	PUNCT
cana-5267	36	13	and	and	CCONJ
cana-5267	36	14	adaptive	adaptive	ADJ
cana-5267	36	15	outcome	outcome	NOUN
cana-5267	36	16	-	-	PUNCT
cana-5267	36	17	determining	determine	VERB
cana-5267	36	18	threshold	threshold	NOUN
cana-5267	36	19	algorithms	algorithm	NOUN
cana-5267	36	20	enabling	enable	VERB
cana-5267	36	21	limitless	limitless	ADJ
cana-5267	36	22	security	security	NOUN
cana-5267	36	23	visibility	visibility	NOUN
cana-5267	36	24	,	,	PUNCT
cana-5267	36	25	real	real	ADJ
cana-5267	36	26	-	-	PUNCT
cana-5267	36	27	time	time	NOUN
cana-5267	36	28	intelligence	intelligence	NOUN
cana-5267	36	29	on	on	ADP
cana-5267	36	30	new	new	ADJ
cana-5267	36	31	risks	risk	NOUN
cana-5267	36	32	,	,	PUNCT
cana-5267	36	33	and	and	CCONJ
cana-5267	36	34	tailored	tailor	VERB
cana-5267	36	35	smart	smart	ADJ
cana-5267	36	36	active	active	ADJ
cana-5267	36	37	risk	risk	NOUN
cana-5267	36	38	management	management	NOUN
cana-5267	36	39	within	within	ADP
cana-5267	36	40	and	and	CCONJ
cana-5267	36	41	outside	outside	ADP
cana-5267	36	42	of	of	ADP
cana-5267	36	43	llm	llm	PROPN
cana-5267	36	44	systems	system	NOUN
cana-5267	36	45	.	.	PUNCT
cana-5267	37	1	fundamentally	fundamentally	ADV
cana-5267	37	2	,	,	PUNCT
cana-5267	37	3	these	these	DET
cana-5267	37	4	tools	tool	NOUN
cana-5267	37	5	enhance	enhance	VERB
cana-5267	37	6	the	the	DET
cana-5267	37	7	atm	atm	NOUN
cana-5267	37	8	apparatus	apparatus	NOUN
cana-5267	37	9	or	or	CCONJ
cana-5267	37	10	machinery	machinery	NOUN
cana-5267	37	11	which	which	PRON
cana-5267	37	12	aims	aim	VERB
cana-5267	37	13	to	to	PART
cana-5267	37	14	refine	refine	VERB
cana-5267	37	15	efficiency	efficiency	NOUN
cana-5267	37	16	,	,	PUNCT
cana-5267	37	17	accuracy	accuracy	NOUN
cana-5267	37	18	,	,	PUNCT
cana-5267	37	19	scope	scope	NOUN
cana-5267	37	20	of	of	ADP
cana-5267	37	21	process	process	NOUN
cana-5267	37	22	automation	automation	NOUN
cana-5267	37	23	,	,	PUNCT
cana-5267	37	24	and	and	CCONJ
cana-5267	37	25	overall	overall	ADJ
cana-5267	37	26	evaluation	evaluation	NOUN
cana-5267	37	27	of	of	ADP
cana-5267	37	28	the	the	DET
cana-5267	37	29	threat	threat	NOUN
cana-5267	37	30	information	information	NOUN
cana-5267	37	31	systems	system	NOUN
cana-5267	37	32	.	.	PUNCT
cana-5267	38	1	hence	hence	ADV
cana-5267	38	2	,	,	PUNCT
cana-5267	38	3	atm	atm	PROPN
cana-5267	38	4	mitigates	mitigate	VERB
cana-5267	38	5	the	the	DET
cana-5267	38	6	burden	burden	NOUN
cana-5267	38	7	which	which	PRON
cana-5267	38	8	cyber	cyber	ADJ
cana-5267	38	9	security	security	NOUN
cana-5267	38	10	arms	arm	NOUN
cana-5267	38	11	bear	bear	VERB
cana-5267	38	12	without	without	ADP
cana-5267	38	13	undermining	undermine	VERB
cana-5267	38	14	the	the	DET
cana-5267	38	15	defense	defense	NOUN
cana-5267	38	16	systems	system	NOUN
cana-5267	38	17	directed	direct	VERB
cana-5267	38	18	at	at	ADP
cana-5267	38	19	attacks	attack	NOUN
cana-5267	38	20	targeting	target	VERB
cana-5267	38	21	the	the	DET
cana-5267	38	22	infrastructures	infrastructure	NOUN
cana-5267	38	23	of	of	ADP
cana-5267	38	24	llm	llm	PROPN
cana-5267	38	25	systems	system	NOUN
cana-5267	38	26	.	.	PUNCT
cana-5267	39	1	communications	communication	NOUN
cana-5267	39	2	on	on	ADP
cana-5267	39	3	applied	apply	VERB
cana-5267	39	4	nonlinear	nonlinear	ADJ
cana-5267	39	5	analysis	analysis	NOUN
cana-5267	39	6	issn	issn	NOUN
cana-5267	39	7	:	:	PUNCT
cana-5267	39	8	1074	1074	NUM
cana-5267	39	9	-	-	PUNCT
cana-5267	39	10	133x	133x	NUM
cana-5267	39	11	vol	vol	VERB
cana-5267	39	12	32	32	NUM
cana-5267	39	13	no	no	NOUN
cana-5267	39	14	.	.	PUNCT
cana-5267	40	1	10s	10	NOUN
cana-5267	40	2	(	(	PUNCT
cana-5267	40	3	2025	2025	NUM
cana-5267	40	4	)	)	PUNCT
cana-5267	40	5	1660	1660	NUM
cana-5267	40	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5267	40	7	attack	attack	NOUN
cana-5267	40	8	vectors	vector	NOUN
cana-5267	40	9	and	and	CCONJ
cana-5267	40	10	their	their	PRON
cana-5267	40	11	ever	ever	ADV
cana-5267	40	12	-	-	PUNCT
cana-5267	40	13	changing	change	VERB
cana-5267	40	14	nature	nature	NOUN
cana-5267	40	15	is	be	AUX
cana-5267	40	16	one	one	NUM
cana-5267	40	17	of	of	ADP
cana-5267	40	18	the	the	DET
cana-5267	40	19	most	most	ADV
cana-5267	40	20	important	important	ADJ
cana-5267	40	21	and	and	CCONJ
cana-5267	40	22	intricate	intricate	ADJ
cana-5267	40	23	challenges	challenge	NOUN
cana-5267	40	24	in	in	ADP
cana-5267	40	25	llm	llm	PROPN
cana-5267	40	26	threat	threat	NOUN
cana-5267	40	27	modeling	modeling	NOUN
cana-5267	40	28	systems	system	NOUN
cana-5267	40	29	.	.	PUNCT
cana-5267	41	1	unlike	unlike	ADP
cana-5267	41	2	software	software	NOUN
cana-5267	41	3	vulnerabilities	vulnerability	NOUN
cana-5267	41	4	that	that	PRON
cana-5267	41	5	arise	arise	VERB
cana-5267	41	6	from	from	ADP
cana-5267	41	7	the	the	DET
cana-5267	41	8	environment	environment	NOUN
cana-5267	41	9	around	around	ADP
cana-5267	41	10	them	they	PRON
cana-5267	41	11	dependent	dependent	ADJ
cana-5267	41	12	on	on	ADP
cana-5267	41	13	a	a	DET
cana-5267	41	14	piece	piece	NOUN
cana-5267	41	15	of	of	ADP
cana-5267	41	16	code	code	NOUN
cana-5267	41	17	,	,	PUNCT
cana-5267	41	18	threats	threat	NOUN
cana-5267	41	19	that	that	PRON
cana-5267	41	20	use	use	VERB
cana-5267	41	21	llm	llm	PROPN
cana-5267	41	22	technology	technology	NOUN
cana-5267	41	23	are	be	AUX
cana-5267	41	24	born	bear	VERB
cana-5267	41	25	from	from	ADP
cana-5267	41	26	dependencies	dependency	NOUN
cana-5267	41	27	related	relate	VERB
cana-5267	41	28	to	to	ADP
cana-5267	41	29	context	context	NOUN
cana-5267	41	30	,	,	PUNCT
cana-5267	41	31	interactions	interaction	NOUN
cana-5267	41	32	with	with	ADP
cana-5267	41	33	prompts	prompt	NOUN
cana-5267	41	34	,	,	PUNCT
cana-5267	41	35	and	and	CCONJ
cana-5267	41	36	the	the	DET
cana-5267	41	37	choreography	choreography	NOUN
cana-5267	41	38	of	of	ADP
cana-5267	41	39	flowing	flow	VERB
cana-5267	41	40	information	information	NOUN
cana-5267	41	41	.	.	PUNCT
cana-5267	42	1	automated	automate	VERB
cana-5267	42	2	threat	threat	NOUN
cana-5267	42	3	modeling	modeling	NOUN
cana-5267	42	4	(	(	PUNCT
cana-5267	42	5	atm	atm	NOUN
cana-5267	42	6	)	)	PUNCT
cana-5267	42	7	solves	solve	VERB
cana-5267	42	8	this	this	DET
cana-5267	42	9	problem	problem	NOUN
cana-5267	42	10	by	by	ADP
cana-5267	42	11	using	use	VERB
cana-5267	42	12	real	real	ADJ
cana-5267	42	13	-	-	PUNCT
cana-5267	42	14	time	time	NOUN
cana-5267	42	15	algorithms	algorithm	NOUN
cana-5267	42	16	for	for	ADP
cana-5267	42	17	threat	threat	NOUN
cana-5267	42	18	detection	detection	NOUN
cana-5267	42	19	,	,	PUNCT
cana-5267	42	20	identifying	identify	VERB
cana-5267	42	21	anomalies	anomaly	NOUN
cana-5267	42	22	,	,	PUNCT
cana-5267	42	23	and	and	CCONJ
cana-5267	42	24	analyzing	analyze	VERB
cana-5267	42	25	potential	potential	ADJ
cana-5267	42	26	risks	risk	NOUN
cana-5267	42	27	using	use	VERB
cana-5267	42	28	ai	ai	NOUN
cana-5267	42	29	-	-	PUNCT
cana-5267	42	30	powered	power	VERB
cana-5267	42	31	adversarial	adversarial	ADJ
cana-5267	42	32	analysis	analysis	NOUN
cana-5267	42	33	systems	system	NOUN
cana-5267	42	34	.	.	PUNCT
cana-5267	43	1	also	also	ADV
cana-5267	43	2	,	,	PUNCT
cana-5267	43	3	ai	ai	VERB
cana-5267	43	4	governance	governance	NOUN
cana-5267	43	5	frameworks	framework	NOUN
cana-5267	43	6	and	and	CCONJ
cana-5267	43	7	ethical	ethical	ADJ
cana-5267	43	8	ai	ai	PROPN
cana-5267	43	9	compliance	compliance	NOUN
cana-5267	43	10	protocols	protocol	NOUN
cana-5267	43	11	focused	focus	VERB
cana-5267	43	12	on	on	ADP
cana-5267	43	13	policy	policy	NOUN
cana-5267	43	14	-	-	PUNCT
cana-5267	43	15	based	base	VERB
cana-5267	43	16	threat	threat	NOUN
cana-5267	43	17	mitigation	mitigation	NOUN
cana-5267	43	18	serve	serve	VERB
cana-5267	43	19	an	an	DET
cana-5267	43	20	equally	equally	ADV
cana-5267	43	21	important	important	ADJ
cana-5267	43	22	purpose	purpose	NOUN
cana-5267	43	23	to	to	PART
cana-5267	43	24	ensure	ensure	VERB
cana-5267	43	25	that	that	SCONJ
cana-5267	43	26	llm	llm	PROPN
cana-5267	43	27	implementations	implementation	NOUN
cana-5267	43	28	are	be	AUX
cana-5267	43	29	monitored	monitor	VERB
cana-5267	43	30	for	for	ADP
cana-5267	43	31	compliance	compliance	NOUN
cana-5267	43	32	with	with	ADP
cana-5267	43	33	regulation	regulation	NOUN
cana-5267	43	34	standards	standard	NOUN
cana-5267	43	35	,	,	PUNCT
cana-5267	43	36	laws	law	NOUN
cana-5267	43	37	pertaining	pertain	VERB
cana-5267	43	38	to	to	ADP
cana-5267	43	39	data	data	NOUN
cana-5267	43	40	privacy	privacy	NOUN
cana-5267	43	41	,	,	PUNCT
cana-5267	43	42	and	and	CCONJ
cana-5267	43	43	cybersecurity	cybersecurity	NOUN
cana-5267	43	44	practices	practice	NOUN
cana-5267	43	45	alongside	alongside	ADP
cana-5267	43	46	other	other	ADJ
cana-5267	43	47	frameworks	framework	NOUN
cana-5267	43	48	minimal	minimal	VERB
cana-5267	43	49	cross	cross	NOUN
cana-5267	43	50	domain	domain	NOUN
cana-5267	43	51	governance	governance	NOUN
cana-5267	43	52	compliance	compliance	NOUN
cana-5267	43	53	interfaces	interface	NOUN
cana-5267	43	54	.	.	PUNCT
cana-5267	44	1	the	the	DET
cana-5267	44	2	use	use	NOUN
cana-5267	44	3	of	of	ADP
cana-5267	44	4	llms	llm	NOUN
cana-5267	44	5	in	in	ADP
cana-5267	44	6	high	high	ADJ
cana-5267	44	7	-	-	PUNCT
cana-5267	44	8	stakes	stake	NOUN
cana-5267	44	9	sectors	sector	NOUN
cana-5267	44	10	like	like	ADP
cana-5267	44	11	banking	banking	NOUN
cana-5267	44	12	,	,	PUNCT
cana-5267	44	13	legal	legal	ADJ
cana-5267	44	14	processes	process	NOUN
cana-5267	44	15	automation	automation	NOUN
cana-5267	44	16	,	,	PUNCT
cana-5267	44	17	military	military	NOUN
cana-5267	44	18	ai	ai	PROPN
cana-5267	44	19	systems	system	NOUN
cana-5267	44	20	,	,	PUNCT
cana-5267	44	21	and	and	CCONJ
cana-5267	44	22	government	government	NOUN
cana-5267	44	23	intelligence	intelligence	NOUN
cana-5267	44	24	has	have	AUX
cana-5267	44	25	increased	increase	VERB
cana-5267	44	26	the	the	DET
cana-5267	44	27	demand	demand	NOUN
cana-5267	44	28	for	for	ADP
cana-5267	44	29	comprehensive	comprehensive	ADJ
cana-5267	44	30	threat	threat	NOUN
cana-5267	44	31	modeling	model	VERB
cana-5267	44	32	strategies	strategy	NOUN
cana-5267	44	33	.	.	PUNCT
cana-5267	45	1	financial	financial	ADJ
cana-5267	45	2	institutions	institution	NOUN
cana-5267	45	3	using	use	VERB
cana-5267	45	4	llm	llm	NOUN
cana-5267	45	5	-	-	PUNCT
cana-5267	45	6	based	base	VERB
cana-5267	45	7	risk	risk	NOUN
cana-5267	45	8	evaluation	evaluation	NOUN
cana-5267	45	9	systems	system	NOUN
cana-5267	45	10	need	need	VERB
cana-5267	45	11	to	to	PART
cana-5267	45	12	be	be	AUX
cana-5267	45	13	sure	sure	ADJ
cana-5267	45	14	that	that	SCONJ
cana-5267	45	15	their	their	PRON
cana-5267	45	16	ai	ai	NOUN
cana-5267	45	17	models	model	NOUN
cana-5267	45	18	are	be	AUX
cana-5267	45	19	protected	protect	VERB
cana-5267	45	20	from	from	ADP
cana-5267	45	21	data	datum	NOUN
cana-5267	45	22	poisoning	poisoning	NOUN
cana-5267	45	23	attacks	attack	NOUN
cana-5267	45	24	,	,	PUNCT
cana-5267	45	25	which	which	PRON
cana-5267	45	26	involve	involve	VERB
cana-5267	45	27	adversarial	adversarial	ADJ
cana-5267	45	28	injection	injection	NOUN
cana-5267	45	29	of	of	ADP
cana-5267	45	30	malicious	malicious	ADJ
cana-5267	45	31	training	training	NOUN
cana-5267	45	32	data	datum	NOUN
cana-5267	45	33	to	to	PART
cana-5267	45	34	skew	skew	VERB
cana-5267	45	35	model	model	NOUN
cana-5267	45	36	predictions	prediction	NOUN
cana-5267	45	37	.	.	PUNCT
cana-5267	46	1	likewise	likewise	ADV
cana-5267	46	2	,	,	PUNCT
cana-5267	46	3	autonomous	autonomous	ADJ
cana-5267	46	4	ai	ai	PROPN
cana-5267	46	5	agents	agent	NOUN
cana-5267	46	6	used	use	VERB
cana-5267	46	7	in	in	ADP
cana-5267	46	8	defense	defense	NOUN
cana-5267	46	9	and	and	CCONJ
cana-5267	46	10	national	national	ADJ
cana-5267	46	11	security	security	NOUN
cana-5267	46	12	plumbing	plumbing	NOUN
cana-5267	46	13	must	must	AUX
cana-5267	46	14	defend	defend	VERB
cana-5267	46	15	themselves	themselves	PRON
cana-5267	46	16	against	against	ADP
cana-5267	46	17	model	model	NOUN
cana-5267	46	18	hijacking	hijacking	NOUN
cana-5267	46	19	exploits	exploit	NOUN
cana-5267	46	20	,	,	PUNCT
cana-5267	46	21	where	where	SCONJ
cana-5267	46	22	an	an	DET
cana-5267	46	23	attacker	attacker	NOUN
cana-5267	46	24	uses	use	VERB
cana-5267	46	25	llm	llm	ADV
cana-5267	46	26	-	-	PUNCT
cana-5267	46	27	supplied	supply	VERB
cana-5267	46	28	results	result	NOUN
cana-5267	46	29	to	to	PART
cana-5267	46	30	spread	spread	VERB
cana-5267	46	31	false	false	ADJ
cana-5267	46	32	information	information	NOUN
cana-5267	46	33	or	or	CCONJ
cana-5267	46	34	carry	carry	VERB
cana-5267	46	35	out	out	ADP
cana-5267	46	36	actions	action	NOUN
cana-5267	46	37	without	without	ADP
cana-5267	46	38	permission	permission	NOUN
cana-5267	46	39	.	.	PUNCT
cana-5267	47	1	in	in	ADP
cana-5267	47	2	these	these	DET
cana-5267	47	3	sensitive	sensitive	ADJ
cana-5267	47	4	areas	area	NOUN
cana-5267	47	5	,	,	PUNCT
cana-5267	47	6	the	the	DET
cana-5267	47	7	inclusion	inclusion	NOUN
cana-5267	47	8	of	of	ADP
cana-5267	47	9	atm	atm	ADJ
cana-5267	47	10	frameworks	framework	NOUN
cana-5267	47	11	improves	improve	VERB
cana-5267	47	12	the	the	DET
cana-5267	47	13	multidisciplinary	multidisciplinary	ADJ
cana-5267	47	14	integration	integration	NOUN
cana-5267	47	15	of	of	ADP
cana-5267	47	16	threat	threat	NOUN
cana-5267	47	17	intelligence	intelligence	NOUN
cana-5267	47	18	,	,	PUNCT
cana-5267	47	19	preparedness	preparedness	NOUN
cana-5267	47	20	for	for	ADP
cana-5267	47	21	incident	incident	NOUN
cana-5267	47	22	response	response	NOUN
cana-5267	47	23	,	,	PUNCT
cana-5267	47	24	predictive	predictive	ADJ
cana-5267	47	25	risk	risk	NOUN
cana-5267	47	26	assessment	assessment	NOUN
cana-5267	47	27	,	,	PUNCT
cana-5267	47	28	and	and	CCONJ
cana-5267	47	29	overall	overall	ADJ
cana-5267	47	30	security	security	NOUN
cana-5267	47	31	posture	posture	NOUN
cana-5267	47	32	of	of	ADP
cana-5267	47	33	systems	system	NOUN
cana-5267	47	34	employing	employ	VERB
cana-5267	47	35	llms	llm	NOUN
cana-5267	47	36	.	.	PUNCT
cana-5267	48	1	another	another	DET
cana-5267	48	2	important	important	ADJ
cana-5267	48	3	component	component	NOUN
cana-5267	48	4	of	of	ADP
cana-5267	48	5	automated	automate	VERB
cana-5267	48	6	threat	threat	NOUN
cana-5267	48	7	modeling	modeling	NOUN
cana-5267	48	8	for	for	ADP
cana-5267	48	9	the	the	DET
cana-5267	48	10	security	security	NOUN
cana-5267	48	11	of	of	ADP
cana-5267	48	12	llms	llm	NOUN
cana-5267	48	13	is	be	AUX
cana-5267	48	14	the	the	DET
cana-5267	48	15	incorporation	incorporation	NOUN
cana-5267	48	16	of	of	ADP
cana-5267	48	17	cyber	cyber	ADJ
cana-5267	48	18	security	security	NOUN
cana-5267	48	19	game	game	NOUN
cana-5267	48	20	-	-	PUNCT
cana-5267	48	21	theory	theory	NOUN
cana-5267	48	22	strategies	strategy	NOUN
cana-5267	48	23	and	and	CCONJ
cana-5267	48	24	ai	ai	VERB
cana-5267	48	25	cyber	cyber	NOUN
cana-5267	48	26	defense	defense	NOUN
cana-5267	48	27	technologies	technology	NOUN
cana-5267	48	28	.	.	PUNCT
cana-5267	49	1	by	by	ADP
cana-5267	49	2	reproducing	reproduce	VERB
cana-5267	49	3	attacks	attack	NOUN
cana-5267	49	4	as	as	ADP
cana-5267	49	5	scenarios	scenario	NOUN
cana-5267	49	6	with	with	ADP
cana-5267	49	7	competing	compete	VERB
cana-5267	49	8	security	security	NOUN
cana-5267	49	9	policies	policy	NOUN
cana-5267	49	10	based	base	VERB
cana-5267	49	11	on	on	ADP
cana-5267	49	12	reinforcement	reinforcement	NOUN
cana-5267	49	13	learning	learning	NOUN
cana-5267	49	14	,	,	PUNCT
cana-5267	49	15	atm	atm	ADJ
cana-5267	49	16	frameworks	framework	NOUN
cana-5267	49	17	can	can	AUX
cana-5267	49	18	alter	alter	VERB
cana-5267	49	19	,	,	PUNCT
cana-5267	49	20	to	to	ADP
cana-5267	49	21	some	some	DET
cana-5267	49	22	degree	degree	NOUN
cana-5267	49	23	,	,	PUNCT
cana-5267	49	24	command	command	NOUN
cana-5267	49	25	hierarchies	hierarchy	NOUN
cana-5267	49	26	for	for	ADP
cana-5267	49	27	automation	automation	NOUN
cana-5267	49	28	workflows	workflow	NOUN
cana-5267	49	29	,	,	PUNCT
cana-5267	49	30	threat	threat	NOUN
cana-5267	49	31	level	level	NOUN
cana-5267	49	32	prioritization	prioritization	NOUN
cana-5267	49	33	,	,	PUNCT
cana-5267	49	34	and	and	CCONJ
cana-5267	49	35	response	response	NOUN
cana-5267	49	36	systems	system	NOUN
cana-5267	49	37	.	.	PUNCT
cana-5267	50	1	for	for	ADP
cana-5267	50	2	instance	instance	NOUN
cana-5267	50	3	,	,	PUNCT
cana-5267	50	4	ai	ai	VERB
cana-5267	50	5	-	-	PUNCT
cana-5267	50	6	assisted	assist	VERB
cana-5267	50	7	attack	attack	NOUN
cana-5267	50	8	surface	surface	NOUN
cana-5267	50	9	mapping	mapping	NOUN
cana-5267	50	10	can	can	AUX
cana-5267	50	11	forecast	forecast	VERB
cana-5267	50	12	pathways	pathway	NOUN
cana-5267	50	13	of	of	ADP
cana-5267	50	14	possible	possible	ADJ
cana-5267	50	15	exploitation	exploitation	NOUN
cana-5267	50	16	by	by	ADP
cana-5267	50	17	examining	examine	VERB
cana-5267	50	18	the	the	DET
cana-5267	50	19	relationships	relationship	NOUN
cana-5267	50	20	between	between	ADP
cana-5267	50	21	system	system	NOUN
cana-5267	50	22	outputs	output	NOUN
cana-5267	50	23	and	and	CCONJ
cana-5267	50	24	inputs	input	NOUN
cana-5267	50	25	,	,	PUNCT
cana-5267	50	26	inter	inter	ADJ
cana-5267	50	27	-	-	ADJ
cana-5267	50	28	system	system	NOUN
cana-5267	50	29	dependencies	dependency	NOUN
cana-5267	50	30	,	,	PUNCT
cana-5267	50	31	and	and	CCONJ
cana-5267	50	32	attack	attack	NOUN
cana-5267	50	33	driving	driving	NOUN
cana-5267	50	34	patterns	pattern	NOUN
cana-5267	50	35	.	.	PUNCT
cana-5267	51	1	moreover	moreover	ADV
cana-5267	51	2	,	,	PUNCT
cana-5267	51	3	the	the	DET
cana-5267	51	4	application	application	NOUN
cana-5267	51	5	of	of	ADP
cana-5267	51	6	federated	federated	ADJ
cana-5267	51	7	learning	learning	NOUN
cana-5267	51	8	in	in	ADP
cana-5267	51	9	the	the	DET
cana-5267	51	10	context	context	NOUN
cana-5267	51	11	of	of	ADP
cana-5267	51	12	llm	llm	PROPN
cana-5267	51	13	security	security	NOUN
cana-5267	51	14	allows	allow	VERB
cana-5267	51	15	decentralized	decentralized	ADJ
cana-5267	51	16	sharing	sharing	NOUN
cana-5267	51	17	of	of	ADP
cana-5267	51	18	intelligence	intelligence	NOUN
cana-5267	51	19	on	on	ADP
cana-5267	51	20	threats	threat	NOUN
cana-5267	51	21	,	,	PUNCT
cana-5267	51	22	enabling	enable	VERB
cana-5267	51	23	organizations	organization	NOUN
cana-5267	51	24	to	to	PART
cana-5267	51	25	pool	pool	VERB
cana-5267	51	26	their	their	PRON
cana-5267	51	27	strengths	strength	NOUN
cana-5267	51	28	for	for	ADP
cana-5267	51	29	defending	defend	VERB
cana-5267	51	30	against	against	ADP
cana-5267	51	31	ai	ai	ADV
cana-5267	51	32	-	-	PUNCT
cana-5267	51	33	driven	drive	VERB
cana-5267	51	34	cyber	cyber	NOUN
cana-5267	51	35	-	-	PUNCT
cana-5267	51	36	sabotage	sabotage	NOUN
cana-5267	51	37	on	on	ADP
cana-5267	51	38	a	a	DET
cana-5267	51	39	worldwide	worldwide	ADJ
cana-5267	51	40	level	level	NOUN
cana-5267	51	41	.	.	PUNCT
cana-5267	52	1	the	the	DET
cana-5267	52	2	ethical	ethical	ADJ
cana-5267	52	3	dimensions	dimension	NOUN
cana-5267	52	4	regarding	regard	VERB
cana-5267	52	5	llm	llm	PROPN
cana-5267	52	6	security	security	NOUN
cana-5267	52	7	further	further	ADJ
cana-5267	52	8	stress	stress	NOUN
cana-5267	52	9	reinforcing	reinforce	VERB
cana-5267	52	10	the	the	DET
cana-5267	52	11	need	need	NOUN
cana-5267	52	12	for	for	ADP
cana-5267	52	13	automated	automate	VERB
cana-5267	52	14	threat	threat	NOUN
cana-5267	52	15	modeling	modeling	NOUN
cana-5267	52	16	for	for	ADP
cana-5267	52	17	integration	integration	NOUN
cana-5267	52	18	into	into	ADP
cana-5267	52	19	ai	ai	PROPN
cana-5267	52	20	systems	system	NOUN
cana-5267	52	21	.	.	PUNCT
cana-5267	53	1	when	when	SCONJ
cana-5267	53	2	llms	llm	NOUN
cana-5267	53	3	gain	gain	VERB
cana-5267	53	4	more	more	ADJ
cana-5267	53	5	control	control	NOUN
cana-5267	53	6	over	over	ADP
cana-5267	53	7	public	public	ADJ
cana-5267	53	8	interactions	interaction	NOUN
cana-5267	53	9	through	through	ADP
cana-5267	53	10	content	content	NOUN
cana-5267	53	11	moderation	moderation	NOUN
cana-5267	53	12	,	,	PUNCT
cana-5267	53	13	automated	automate	VERB
cana-5267	53	14	legal	legal	ADJ
cana-5267	53	15	services	service	NOUN
cana-5267	53	16	,	,	PUNCT
cana-5267	53	17	or	or	CCONJ
cana-5267	53	18	even	even	ADV
cana-5267	53	19	providing	provide	VERB
cana-5267	53	20	medical	medical	ADJ
cana-5267	53	21	diagnoses	diagnosis	NOUN
cana-5267	53	22	,	,	PUNCT
cana-5267	53	23	they	they	PRON
cana-5267	53	24	become	become	VERB
cana-5267	53	25	amplifiers	amplifier	NOUN
cana-5267	53	26	for	for	ADP
cana-5267	53	27	bias	bias	NOUN
cana-5267	53	28	.	.	PUNCT
cana-5267	54	1	automation	automation	NOUN
cana-5267	54	2	of	of	ADP
cana-5267	54	3	fairness	fairness	NOUN
cana-5267	54	4	assessment	assessment	NOUN
cana-5267	54	5	and	and	CCONJ
cana-5267	54	6	governance	governance	NOUN
cana-5267	54	7	requires	require	VERB
cana-5267	54	8	atm	atm	NOUN
cana-5267	54	9	methodologies	methodology	NOUN
cana-5267	54	10	.	.	PUNCT
cana-5267	55	1	such	such	ADJ
cana-5267	55	2	methodologies	methodology	NOUN
cana-5267	55	3	need	need	VERB
cana-5267	55	4	to	to	PART
cana-5267	55	5	embed	embed	VERB
cana-5267	55	6	algorithms	algorithm	NOUN
cana-5267	55	7	to	to	PART
cana-5267	55	8	detect	detect	VERB
cana-5267	55	9	biases	bias	NOUN
cana-5267	55	10	,	,	PUNCT
cana-5267	55	11	adversarial	adversarial	ADJ
cana-5267	55	12	anti	anti	ADJ
cana-5267	55	13	-	-	ADJ
cana-5267	55	14	debiasing	debiasing	ADJ
cana-5267	55	15	approaches	approach	NOUN
cana-5267	55	16	,	,	PUNCT
cana-5267	55	17	and	and	CCONJ
cana-5267	55	18	policies	policy	NOUN
cana-5267	55	19	to	to	PART
cana-5267	55	20	mitigate	mitigate	VERB
cana-5267	55	21	decision	decision	NOUN
cana-5267	55	22	-	-	PUNCT
cana-5267	55	23	making	make	VERB
cana-5267	55	24	misinformation	misinformation	NOUN
cana-5267	55	25	that	that	PRON
cana-5267	55	26	endorses	endorse	VERB
cana-5267	55	27	unfair	unfair	ADJ
cana-5267	55	28	stereotypes	stereotype	NOUN
cana-5267	55	29	and	and	CCONJ
cana-5267	55	30	unethical	unethical	ADJ
cana-5267	55	31	outcomes	outcome	NOUN
cana-5267	55	32	.	.	PUNCT
cana-5267	56	1	these	these	DET
cana-5267	56	2	efforts	effort	NOUN
cana-5267	56	3	are	be	AUX
cana-5267	56	4	in	in	ADP
cana-5267	56	5	line	line	NOUN
cana-5267	56	6	with	with	ADP
cana-5267	56	7	establishing	establish	VERB
cana-5267	56	8	ai	ai	NOUN
cana-5267	56	9	safety	safety	NOUN
cana-5267	56	10	regulations	regulation	NOUN
cana-5267	56	11	frameworks	framework	NOUN
cana-5267	56	12	along	along	ADP
cana-5267	56	13	with	with	ADP
cana-5267	56	14	responsible	responsible	ADJ
cana-5267	56	15	ai	ai	AUX
cana-5267	56	16	endorsing	endorse	VERB
cana-5267	56	17	regulations	regulation	NOUN
cana-5267	56	18	to	to	PART
cana-5267	56	19	be	be	AUX
cana-5267	56	20	actively	actively	ADV
cana-5267	56	21	enforced	enforce	VERB
cana-5267	56	22	like	like	ADP
cana-5267	56	23	the	the	DET
cana-5267	56	24	eu	eu	PROPN
cana-5267	56	25	ai	ai	PROPN
cana-5267	56	26	act	act	PROPN
cana-5267	56	27	and	and	CCONJ
cana-5267	56	28	nist	nist	NOUN
cana-5267	56	29	ai	ai	VERB
cana-5267	56	30	risk	risk	NOUN
cana-5267	56	31	management	management	NOUN
cana-5267	56	32	framework	framework	NOUN
cana-5267	56	33	.	.	PUNCT
cana-5267	57	1	as	as	SCONJ
cana-5267	57	2	llms	llm	NOUN
cana-5267	57	3	are	be	AUX
cana-5267	57	4	adopted	adopt	VERB
cana-5267	57	5	in	in	ADP
cana-5267	57	6	different	different	ADJ
cana-5267	57	7	sectors	sector	NOUN
cana-5267	57	8	ai	ai	AUX
cana-5267	57	9	security	security	NOUN
cana-5267	57	10	compliance	compliance	NOUN
cana-5267	57	11	is	be	AUX
cana-5267	57	12	coming	come	VERB
cana-5267	57	13	under	under	ADP
cana-5267	57	14	scrutiny	scrutiny	NOUN
cana-5267	57	15	for	for	ADP
cana-5267	57	16	legal	legal	ADJ
cana-5267	57	17	responsibility	responsibility	NOUN
cana-5267	57	18	and	and	CCONJ
cana-5267	57	19	regulatory	regulatory	ADJ
cana-5267	57	20	practices	practice	NOUN
cana-5267	57	21	.	.	PUNCT
cana-5267	58	1	compliance	compliance	NOUN
cana-5267	58	2	frameworks	framework	NOUN
cana-5267	58	3	for	for	ADP
cana-5267	58	4	the	the	DET
cana-5267	58	5	use	use	NOUN
cana-5267	58	6	of	of	ADP
cana-5267	58	7	ai	ai	PRON
cana-5267	58	8	technology	technology	NOUN
cana-5267	58	9	are	be	AUX
cana-5267	58	10	being	be	AUX
cana-5267	58	11	legislated	legislate	VERB
cana-5267	58	12	like	like	ADP
cana-5267	58	13	iso	iso	NOUN
cana-5267	58	14	/	/	SYM
cana-5267	58	15	iec	iec	PROPN
cana-5267	58	16	27001	27001	NUM
cana-5267	58	17	concerning	concern	VERB
cana-5267	58	18	ai	ai	INTJ
cana-5267	58	19	risk	risk	NOUN
cana-5267	58	20	management	management	NOUN
cana-5267	58	21	,	,	PUNCT
cana-5267	58	22	gdpr	gdpr	NOUN
cana-5267	58	23	concerning	concern	VERB
cana-5267	58	24	ai	ai	PROPN
cana-5267	58	25	data	data	NOUN
cana-5267	58	26	communications	communication	NOUN
cana-5267	58	27	on	on	ADP
cana-5267	58	28	applied	apply	VERB
cana-5267	58	29	nonlinear	nonlinear	ADJ
cana-5267	58	30	analysis	analysis	NOUN
cana-5267	58	31	issn	issn	NOUN
cana-5267	58	32	:	:	PUNCT
cana-5267	58	33	1074	1074	NUM
cana-5267	58	34	-	-	PUNCT
cana-5267	58	35	133x	133x	NUM
cana-5267	58	36	vol	vol	VERB
cana-5267	58	37	32	32	NUM
cana-5267	58	38	no	no	NOUN
cana-5267	58	39	.	.	PUNCT
cana-5267	59	1	10s	10	NOUN
cana-5267	59	2	(	(	PUNCT
cana-5267	59	3	2025	2025	NUM
cana-5267	59	4	)	)	PUNCT
cana-5267	59	5	1661	1661	NUM
cana-5267	59	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5267	59	7	protection	protection	NOUN
cana-5267	59	8	,	,	PUNCT
cana-5267	59	9	and	and	CCONJ
cana-5267	59	10	soc	soc	NOUN
cana-5267	59	11	2	2	NUM
cana-5267	59	12	regarding	regard	VERB
cana-5267	59	13	ai	ai	INTJ
cana-5267	59	14	security	security	NOUN
cana-5267	59	15	auditing	auditing	NOUN
cana-5267	59	16	.	.	PUNCT
cana-5267	60	1	automated	automate	VERB
cana-5267	60	2	threat	threat	NOUN
cana-5267	60	3	modeling	model	VERB
cana-5267	60	4	frameworks	framework	NOUN
cana-5267	60	5	can	can	AUX
cana-5267	60	6	facilitate	facilitate	VERB
cana-5267	60	7	compliance	compliance	NOUN
cana-5267	60	8	by	by	ADP
cana-5267	60	9	incorporating	incorporate	VERB
cana-5267	60	10	security	security	NOUN
cana-5267	60	11	control	control	NOUN
cana-5267	60	12	implementation	implementation	NOUN
cana-5267	60	13	,	,	PUNCT
cana-5267	60	14	automated	automate	VERB
cana-5267	60	15	compliance	compliance	NOUN
cana-5267	60	16	monitoring	monitoring	NOUN
cana-5267	60	17	,	,	PUNCT
cana-5267	60	18	and	and	CCONJ
cana-5267	60	19	continuous	continuous	ADJ
cana-5267	60	20	audit	audit	NOUN
cana-5267	60	21	log	log	NOUN
cana-5267	60	22	generation	generation	NOUN
cana-5267	60	23	capturing	capture	VERB
cana-5267	60	24	evidence	evidence	NOUN
cana-5267	60	25	of	of	ADP
cana-5267	60	26	compliance	compliance	NOUN
cana-5267	60	27	into	into	ADP
cana-5267	60	28	system	system	NOUN
cana-5267	60	29	deployment	deployment	NOUN
cana-5267	60	30	pipelines	pipeline	NOUN
cana-5267	60	31	.	.	PUNCT
cana-5267	61	1	these	these	DET
cana-5267	61	2	measures	measure	NOUN
cana-5267	61	3	help	help	VERB
cana-5267	61	4	ensure	ensure	VERB
cana-5267	61	5	that	that	SCONJ
cana-5267	61	6	such	such	ADJ
cana-5267	61	7	systems	system	NOUN
cana-5267	61	8	operate	operate	VERB
cana-5267	61	9	within	within	ADP
cana-5267	61	10	defined	define	VERB
cana-5267	61	11	legal	legal	ADJ
cana-5267	61	12	,	,	PUNCT
cana-5267	61	13	ethical	ethical	ADJ
cana-5267	61	14	,	,	PUNCT
cana-5267	61	15	and	and	CCONJ
cana-5267	61	16	domain	domain	NOUN
cana-5267	61	17	-	-	PUNCT
cana-5267	61	18	specific	specific	ADJ
cana-5267	61	19	boundaries	boundary	NOUN
cana-5267	61	20	while	while	SCONJ
cana-5267	61	21	minimizing	minimize	VERB
cana-5267	61	22	exposure	exposure	NOUN
cana-5267	61	23	to	to	ADP
cana-5267	61	24	the	the	DET
cana-5267	61	25	legal	legal	ADJ
cana-5267	61	26	implications	implication	NOUN
cana-5267	61	27	of	of	ADP
cana-5267	61	28	ai	ai	VERB
cana-5267	61	29	security	security	NOUN
cana-5267	61	30	breaches	breach	NOUN
cana-5267	61	31	,	,	PUNCT
cana-5267	61	32	privacy	privacy	NOUN
cana-5267	61	33	infringements	infringement	NOUN
cana-5267	61	34	,	,	PUNCT
cana-5267	61	35	and	and	CCONJ
cana-5267	61	36	the	the	DET
cana-5267	61	37	accountability	accountability	NOUN
cana-5267	61	38	gaps	gap	NOUN
cana-5267	61	39	of	of	ADP
cana-5267	61	40	algorithmic	algorithmic	ADJ
cana-5267	61	41	decision	decision	NOUN
cana-5267	61	42	-	-	PUNCT
cana-5267	61	43	making	making	NOUN
cana-5267	61	44	.	.	PUNCT
cana-5267	62	1	even	even	ADV
cana-5267	62	2	with	with	ADP
cana-5267	62	3	the	the	DET
cana-5267	62	4	progress	progress	NOUN
cana-5267	62	5	made	make	VERB
cana-5267	62	6	towards	towards	ADP
cana-5267	62	7	automated	automate	VERB
cana-5267	62	8	threat	threat	NOUN
cana-5267	62	9	modeling	modeling	NOUN
cana-5267	62	10	for	for	ADP
cana-5267	62	11	llm	llm	PROPN
cana-5267	62	12	system	system	NOUN
cana-5267	62	13	security	security	NOUN
cana-5267	62	14	,	,	PUNCT
cana-5267	62	15	there	there	PRON
cana-5267	62	16	are	be	VERB
cana-5267	62	17	still	still	ADV
cana-5267	62	18	some	some	DET
cana-5267	62	19	challenges	challenge	NOUN
cana-5267	62	20	left	leave	VERB
cana-5267	62	21	for	for	SCONJ
cana-5267	62	22	researchers	researcher	NOUN
cana-5267	62	23	to	to	PART
cana-5267	62	24	tackle	tackle	VERB
cana-5267	62	25	.	.	PUNCT
cana-5267	63	1	one	one	NUM
cana-5267	63	2	of	of	ADP
cana-5267	63	3	the	the	DET
cana-5267	63	4	foremost	foremost	ADJ
cana-5267	63	5	obstacles	obstacle	NOUN
cana-5267	63	6	is	be	AUX
cana-5267	63	7	how	how	SCONJ
cana-5267	63	8	effective	effective	ADJ
cana-5267	63	9	atm	atm	ADJ
cana-5267	63	10	frameworks	framework	NOUN
cana-5267	63	11	are	be	AUX
cana-5267	63	12	for	for	ADP
cana-5267	63	13	the	the	DET
cana-5267	63	14	security	security	NOUN
cana-5267	63	15	of	of	ADP
cana-5267	63	16	systems	system	NOUN
cana-5267	63	17	with	with	ADP
cana-5267	63	18	numerous	numerous	ADJ
cana-5267	63	19	ai	ai	NOUN
cana-5267	63	20	models	model	NOUN
cana-5267	63	21	operating	operate	VERB
cana-5267	63	22	in	in	ADP
cana-5267	63	23	different	different	ADJ
cana-5267	63	24	cloud	cloud	NOUN
cana-5267	63	25	environments	environment	NOUN
cana-5267	63	26	.	.	PUNCT
cana-5267	64	1	there	there	PRON
cana-5267	64	2	is	be	VERB
cana-5267	64	3	still	still	ADV
cana-5267	64	4	work	work	NOUN
cana-5267	64	5	needed	need	VERB
cana-5267	64	6	to	to	PART
cana-5267	64	7	be	be	AUX
cana-5267	64	8	done	do	VERB
cana-5267	64	9	regarding	regard	VERB
cana-5267	64	10	the	the	DET
cana-5267	64	11	low	low	ADJ
cana-5267	64	12	-	-	PUNCT
cana-5267	64	13	latency	latency	NOUN
cana-5267	64	14	,	,	PUNCT
cana-5267	64	15	real	real	ADJ
cana-5267	64	16	-	-	PUNCT
cana-5267	64	17	time	time	NOUN
cana-5267	64	18	threat	threat	NOUN
cana-5267	64	19	detection	detection	NOUN
cana-5267	64	20	,	,	PUNCT
cana-5267	64	21	model	model	NOUN
cana-5267	64	22	performance	performance	NOUN
cana-5267	64	23	,	,	PUNCT
cana-5267	64	24	computational	computational	ADJ
cana-5267	64	25	efficiency	efficiency	NOUN
cana-5267	64	26	,	,	PUNCT
cana-5267	64	27	and	and	CCONJ
cana-5267	64	28	user	user	NOUN
cana-5267	64	29	experience	experience	NOUN
cana-5267	64	30	balance	balance	NOUN
cana-5267	64	31	.	.	PUNCT
cana-5267	65	1	moreover	moreover	ADV
cana-5267	65	2	,	,	PUNCT
cana-5267	65	3	zero	zero	NUM
cana-5267	65	4	-	-	PUNCT
cana-5267	65	5	day	day	NOUN
cana-5267	65	6	llm	llm	NOUN
cana-5267	65	7	exploitation	exploitation	NOUN
cana-5267	65	8	attacks	attack	NOUN
cana-5267	65	9	add	add	VERB
cana-5267	65	10	to	to	ADP
cana-5267	65	11	security	security	NOUN
cana-5267	65	12	concerns	concern	NOUN
cana-5267	65	13	,	,	PUNCT
cana-5267	65	14	making	make	VERB
cana-5267	65	15	the	the	DET
cana-5267	65	16	need	need	NOUN
cana-5267	65	17	for	for	SCONJ
cana-5267	65	18	ai	ai	AUX
cana-5267	65	19	powered	power	VERB
cana-5267	65	20	adaptive	adaptive	ADJ
cana-5267	65	21	active	active	ADJ
cana-5267	65	22	security	security	NOUN
cana-5267	65	23	systems	system	NOUN
cana-5267	65	24	that	that	PRON
cana-5267	65	25	employ	employ	VERB
cana-5267	65	26	continuous	continuous	ADJ
cana-5267	65	27	adversarial	adversarial	ADJ
cana-5267	65	28	training	training	NOUN
cana-5267	65	29	and	and	CCONJ
cana-5267	65	30	real	real	ADJ
cana-5267	65	31	-	-	PUNCT
cana-5267	65	32	time	time	NOUN
cana-5267	65	33	anomaly	anomaly	NOUN
cana-5267	65	34	defenses	defense	NOUN
cana-5267	65	35	necessary	necessary	ADJ
cana-5267	65	36	to	to	PART
cana-5267	65	37	counter	counter	VERB
cana-5267	65	38	these	these	DET
cana-5267	65	39	threats	threat	NOUN
cana-5267	65	40	promptly	promptly	ADV
cana-5267	65	41	.	.	PUNCT
cana-5267	66	1	this	this	DET
cana-5267	66	2	work	work	NOUN
cana-5267	66	3	seeks	seek	VERB
cana-5267	66	4	to	to	PART
cana-5267	66	5	remedy	remedy	VERB
cana-5267	66	6	the	the	DET
cana-5267	66	7	imbalance	imbalance	NOUN
cana-5267	66	8	between	between	ADP
cana-5267	66	9	ai	ai	PRON
cana-5267	66	10	driven	drive	VERB
cana-5267	66	11	security	security	NOUN
cana-5267	66	12	and	and	CCONJ
cana-5267	66	13	automated	automate	VERB
cana-5267	66	14	risk	risk	NOUN
cana-5267	66	15	evaluation	evaluation	NOUN
cana-5267	66	16	strategies	strategy	NOUN
cana-5267	66	17	by	by	ADP
cana-5267	66	18	designing	design	VERB
cana-5267	66	19	an	an	DET
cana-5267	66	20	integrated	integrate	VERB
cana-5267	66	21	llm	llm	NOUN
cana-5267	66	22	system	system	NOUN
cana-5267	66	23	focused	focus	VERB
cana-5267	66	24	on	on	ADP
cana-5267	66	25	agile	agile	ADJ
cana-5267	66	26	frameworks	framework	NOUN
cana-5267	66	27	for	for	ADP
cana-5267	66	28	atm	atm	NOUN
cana-5267	66	29	driven	drive	VERB
cana-5267	66	30	threat	threat	NOUN
cana-5267	66	31	modeling	modeling	NOUN
cana-5267	66	32	.	.	PUNCT
cana-5267	67	1	through	through	ADP
cana-5267	67	2	high	high	ADJ
cana-5267	67	3	-	-	PUNCT
cana-5267	67	4	level	level	NOUN
cana-5267	67	5	cyber	cyber	NOUN
cana-5267	67	6	defense	defense	NOUN
cana-5267	67	7	analytics	analytic	NOUN
cana-5267	67	8	,	,	PUNCT
cana-5267	67	9	resiliency	resiliency	NOUN
cana-5267	67	10	,	,	PUNCT
cana-5267	67	11	and	and	CCONJ
cana-5267	67	12	automation	automation	NOUN
cana-5267	67	13	,	,	PUNCT
cana-5267	67	14	the	the	DET
cana-5267	67	15	proposed	propose	VERB
cana-5267	67	16	framework	framework	NOUN
cana-5267	67	17	provides	provide	VERB
cana-5267	67	18	next	next	ADJ
cana-5267	67	19	generation	generation	NOUN
cana-5267	67	20	ai	ai	VERB
cana-5267	67	21	applications	application	NOUN
cana-5267	67	22	advanced	advanced	ADJ
cana-5267	67	23	control	control	NOUN
cana-5267	67	24	and	and	CCONJ
cana-5267	67	25	reduced	reduce	VERB
cana-5267	67	26	threat	threat	NOUN
cana-5267	67	27	response	response	NOUN
cana-5267	67	28	time	time	NOUN
cana-5267	67	29	.	.	PUNCT
cana-5267	68	1	these	these	DET
cana-5267	68	2	objectives	objective	NOUN
cana-5267	68	3	enable	enable	VERB
cana-5267	68	4	organizations	organization	NOUN
cana-5267	68	5	to	to	PART
cana-5267	68	6	incorporate	incorporate	VERB
cana-5267	68	7	large	large	ADJ
cana-5267	68	8	language	language	NOUN
cana-5267	68	9	model	model	NOUN
cana-5267	68	10	systems	system	NOUN
cana-5267	68	11	seamlessly	seamlessly	ADV
cana-5267	68	12	,	,	PUNCT
cana-5267	68	13	securely	securely	ADV
cana-5267	68	14	and	and	CCONJ
cana-5267	68	15	responsibly	responsibly	ADV
cana-5267	68	16	while	while	SCONJ
cana-5267	68	17	contributing	contribute	VERB
cana-5267	68	18	to	to	ADP
cana-5267	68	19	the	the	DET
cana-5267	68	20	growing	grow	VERB
cana-5267	68	21	body	body	NOUN
cana-5267	68	22	of	of	ADP
cana-5267	68	23	ai	ai	PROPN
cana-5267	68	24	security	security	NOUN
cana-5267	68	25	research	research	NOUN
cana-5267	68	26	that	that	PRON
cana-5267	68	27	focuses	focus	VERB
cana-5267	68	28	on	on	ADP
cana-5267	68	29	automated	automate	VERB
cana-5267	68	30	threat	threat	NOUN
cana-5267	68	31	modeling	modeling	NOUN
cana-5267	68	32	.	.	PUNCT
cana-5267	69	1	literature	literature	NOUN
cana-5267	69	2	survey	survey	VERB
cana-5267	69	3	the	the	DET
cana-5267	69	4	development	development	NOUN
cana-5267	69	5	of	of	ADP
cana-5267	69	6	large	large	ADJ
cana-5267	69	7	language	language	NOUN
cana-5267	69	8	models	model	NOUN
cana-5267	69	9	(	(	PUNCT
cana-5267	69	10	llms	llm	NOUN
cana-5267	69	11	)	)	PUNCT
cana-5267	69	12	has	have	AUX
cana-5267	69	13	brought	bring	VERB
cana-5267	69	14	new	new	ADJ
cana-5267	69	15	capabilities	capability	NOUN
cana-5267	69	16	in	in	ADP
cana-5267	69	17	applications	application	NOUN
cana-5267	69	18	,	,	PUNCT
cana-5267	69	19	artificial	artificial	ADJ
cana-5267	69	20	intelligence	intelligence	NOUN
cana-5267	69	21	(	(	PUNCT
cana-5267	69	22	ai	ai	NOUN
cana-5267	69	23	)	)	PUNCT
cana-5267	69	24	,	,	PUNCT
cana-5267	69	25	and	and	CCONJ
cana-5267	69	26	even	even	ADV
cana-5267	69	27	machine	machine	NOUN
cana-5267	69	28	learning	learning	NOUN
cana-5267	69	29	(	(	PUNCT
cana-5267	69	30	ml	ml	NOUN
cana-5267	69	31	)	)	PUNCT
cana-5267	69	32	.	.	PUNCT
cana-5267	70	1	nevertheless	nevertheless	ADV
cana-5267	70	2	,	,	PUNCT
cana-5267	70	3	the	the	DET
cana-5267	70	4	widespread	widespread	ADJ
cana-5267	70	5	usage	usage	NOUN
cana-5267	70	6	of	of	ADP
cana-5267	70	7	these	these	DET
cana-5267	70	8	models	model	NOUN
cana-5267	70	9	comes	come	VERB
cana-5267	70	10	with	with	ADP
cana-5267	70	11	new	new	ADJ
cana-5267	70	12	security	security	NOUN
cana-5267	70	13	challenges	challenge	NOUN
cana-5267	70	14	and	and	CCONJ
cana-5267	70	15	risks	risk	NOUN
cana-5267	70	16	that	that	PRON
cana-5267	70	17	need	need	VERB
cana-5267	70	18	to	to	PART
cana-5267	70	19	be	be	AUX
cana-5267	70	20	managed	manage	VERB
cana-5267	70	21	in	in	ADP
cana-5267	70	22	a	a	DET
cana-5267	70	23	structured	structured	ADJ
cana-5267	70	24	way	way	NOUN
cana-5267	70	25	.	.	PUNCT
cana-5267	71	1	automated	automate	VERB
cana-5267	71	2	threat	threat	NOUN
cana-5267	71	3	modeling	modeling	NOUN
cana-5267	71	4	(	(	PUNCT
cana-5267	71	5	atm	atm	NOUN
cana-5267	71	6	)	)	PUNCT
cana-5267	71	7	has	have	AUX
cana-5267	71	8	become	become	VERB
cana-5267	71	9	one	one	NUM
cana-5267	71	10	of	of	ADP
cana-5267	71	11	the	the	DET
cana-5267	71	12	most	most	ADV
cana-5267	71	13	important	important	ADJ
cana-5267	71	14	techniques	technique	NOUN
cana-5267	71	15	for	for	ADP
cana-5267	71	16	identifying	identify	VERB
cana-5267	71	17	,	,	PUNCT
cana-5267	71	18	reducing	reduce	VERB
cana-5267	71	19	,	,	PUNCT
cana-5267	71	20	and	and	CCONJ
cana-5267	71	21	preventing	prevent	VERB
cana-5267	71	22	security	security	NOUN
cana-5267	71	23	risks	risk	NOUN
cana-5267	71	24	during	during	ADP
cana-5267	71	25	llm	llm	PROPN
cana-5267	71	26	system	system	NOUN
cana-5267	71	27	integration	integration	NOUN
cana-5267	71	28	.	.	PUNCT
cana-5267	72	1	scholars	scholar	NOUN
cana-5267	72	2	have	have	AUX
cana-5267	72	3	studied	study	VERB
cana-5267	72	4	a	a	DET
cana-5267	72	5	number	number	NOUN
cana-5267	72	6	of	of	ADP
cana-5267	72	7	machine	machine	NOUN
cana-5267	72	8	learning	learning	NOUN
cana-5267	72	9	methods	method	NOUN
cana-5267	72	10	,	,	PUNCT
cana-5267	72	11	cybersecurity	cybersecurity	NOUN
cana-5267	72	12	frameworks	framework	NOUN
cana-5267	72	13	,	,	PUNCT
cana-5267	72	14	and	and	CCONJ
cana-5267	72	15	nlp	nlp	NOUN
cana-5267	72	16	-	-	PUNCT
cana-5267	72	17	based	base	VERB
cana-5267	72	18	threat	threat	NOUN
cana-5267	72	19	monitoring	monitor	VERB
cana-5267	72	20	systems	system	NOUN
cana-5267	72	21	designed	design	VERB
cana-5267	72	22	to	to	PART
cana-5267	72	23	increase	increase	VERB
cana-5267	72	24	the	the	DET
cana-5267	72	25	robustness	robustness	NOUN
cana-5267	72	26	of	of	ADP
cana-5267	72	27	llm	llm	NOUN
cana-5267	72	28	-	-	PUNCT
cana-5267	72	29	based	base	VERB
cana-5267	72	30	systems	system	NOUN
cana-5267	72	31	against	against	ADP
cana-5267	72	32	adversarial	adversarial	ADJ
cana-5267	72	33	control	control	NOUN
cana-5267	72	34	,	,	PUNCT
cana-5267	72	35	data	data	NOUN
cana-5267	72	36	poisoning	poisoning	NOUN
cana-5267	72	37	,	,	PUNCT
cana-5267	72	38	model	model	NOUN
cana-5267	72	39	inversion	inversion	NOUN
cana-5267	72	40	,	,	PUNCT
cana-5267	72	41	and	and	CCONJ
cana-5267	72	42	unauthorized	unauthorized	ADJ
cana-5267	72	43	access	access	NOUN
cana-5267	72	44	.	.	PUNCT
cana-5267	73	1	this	this	DET
cana-5267	73	2	literature	literature	NOUN
cana-5267	73	3	summary	summary	NOUN
cana-5267	73	4	focuses	focus	VERB
cana-5267	73	5	on	on	ADP
cana-5267	73	6	recent	recent	ADJ
cana-5267	73	7	studies	study	NOUN
cana-5267	73	8	on	on	ADP
cana-5267	73	9	ai	ai	NOUN
cana-5267	73	10	-	-	PUNCT
cana-5267	73	11	based	base	VERB
cana-5267	73	12	threat	threat	NOUN
cana-5267	73	13	modeling	modeling	NOUN
cana-5267	73	14	,	,	PUNCT
cana-5267	73	15	adversarial	adversarial	ADJ
cana-5267	73	16	machine	machine	NOUN
cana-5267	73	17	learning	learning	NOUN
cana-5267	73	18	,	,	PUNCT
cana-5267	73	19	risk	risk	NOUN
cana-5267	73	20	assessment	assessment	NOUN
cana-5267	73	21	frameworks	framework	NOUN
cana-5267	73	22	,	,	PUNCT
cana-5267	73	23	and	and	CCONJ
cana-5267	73	24	automated	automate	VERB
cana-5267	73	25	cybersecurity	cybersecurity	NOUN
cana-5267	73	26	specific	specific	ADJ
cana-5267	73	27	to	to	ADP
cana-5267	73	28	llm	llm	ADV
cana-5267	73	29	-	-	PUNCT
cana-5267	73	30	based	base	VERB
cana-5267	73	31	systems	system	NOUN
cana-5267	73	32	.	.	PUNCT
cana-5267	74	1	a	a	DET
cana-5267	74	2	few	few	ADJ
cana-5267	74	3	of	of	ADP
cana-5267	74	4	them	they	PRON
cana-5267	74	5	attempted	attempt	VERB
cana-5267	74	6	speeding	speed	VERB
cana-5267	74	7	the	the	DET
cana-5267	74	8	process	process	NOUN
cana-5267	74	9	of	of	ADP
cana-5267	74	10	detecting	detect	VERB
cana-5267	74	11	threats	threat	NOUN
cana-5267	74	12	in	in	ADP
cana-5267	74	13	ai	ai	PROPN
cana-5267	74	14	systems	system	NOUN
cana-5267	74	15	using	use	VERB
cana-5267	74	16	hybrid	hybrid	NOUN
cana-5267	74	17	ml	ml	NOUN
cana-5267	74	18	approaches	approach	NOUN
cana-5267	74	19	.	.	PUNCT
cana-5267	75	1	an	an	DET
cana-5267	75	2	example	example	NOUN
cana-5267	75	3	of	of	ADP
cana-5267	75	4	this	this	DET
cana-5267	75	5	work	work	NOUN
cana-5267	75	6	is	be	AUX
cana-5267	75	7	brown	brown	ADJ
cana-5267	75	8	and	and	CCONJ
cana-5267	75	9	johnson	johnson	PROPN
cana-5267	75	10	's	's	PART
cana-5267	75	11	[	[	X
cana-5267	75	12	1	1	NUM
cana-5267	75	13	]	]	PUNCT
cana-5267	75	14	which	which	PRON
cana-5267	75	15	developed	develop	VERB
cana-5267	75	16	an	an	DET
cana-5267	75	17	ai	ai	NOUN
cana-5267	75	18	threat	threat	NOUN
cana-5267	75	19	modeling	model	VERB
cana-5267	75	20	framework	framework	NOUN
cana-5267	75	21	that	that	PRON
cana-5267	75	22	evaluates	evaluate	VERB
cana-5267	75	23	possible	possible	ADJ
cana-5267	75	24	llm	llm	PROPN
cana-5267	75	25	applications	application	NOUN
cana-5267	75	26	vulnerabilities	vulnerability	NOUN
cana-5267	75	27	.	.	PUNCT
cana-5267	76	1	their	their	PRON
cana-5267	76	2	findings	finding	NOUN
cana-5267	76	3	show	show	VERB
cana-5267	76	4	how	how	SCONJ
cana-5267	76	5	automated	automate	VERB
cana-5267	76	6	risk	risk	NOUN
cana-5267	76	7	assessment	assessment	NOUN
cana-5267	76	8	processes	process	NOUN
cana-5267	76	9	combined	combine	VERB
cana-5267	76	10	with	with	ADP
cana-5267	76	11	anomaly	anomaly	NOUN
cana-5267	76	12	detection	detection	NOUN
cana-5267	76	13	and	and	CCONJ
cana-5267	76	14	monitoring	monitoring	NOUN
cana-5267	76	15	might	might	AUX
cana-5267	76	16	aid	aid	VERB
cana-5267	76	17	securing	secure	VERB
cana-5267	76	18	llms	llm	NOUN
cana-5267	76	19	.	.	PUNCT
cana-5267	77	1	gupta	gupta	PROPN
cana-5267	77	2	et	et	PROPN
cana-5267	77	3	al	al	PROPN
cana-5267	77	4	.	.	PUNCT
cana-5267	78	1	[	[	X
cana-5267	78	2	2	2	X
cana-5267	78	3	]	]	PUNCT
cana-5267	78	4	have	have	AUX
cana-5267	78	5	also	also	ADV
cana-5267	78	6	conducted	conduct	VERB
cana-5267	78	7	work	work	NOUN
cana-5267	78	8	wherein	wherein	SCONJ
cana-5267	78	9	a	a	DET
cana-5267	78	10	machine	machine	NOUN
cana-5267	78	11	learning	learn	VERB
cana-5267	78	12	model	model	NOUN
cana-5267	78	13	for	for	ADP
cana-5267	78	14	cybersecurity	cybersecurity	NOUN
cana-5267	78	15	is	be	AUX
cana-5267	78	16	developed	develop	VERB
cana-5267	78	17	that	that	PRON
cana-5267	78	18	uses	use	VERB
cana-5267	78	19	predictive	predictive	ADJ
cana-5267	78	20	intelligence	intelligence	NOUN
cana-5267	78	21	,	,	PUNCT
cana-5267	78	22	behavioral	behavioral	NOUN
cana-5267	78	23	,	,	PUNCT
cana-5267	78	24	and	and	CCONJ
cana-5267	78	25	attack	attack	NOUN
cana-5267	78	26	profile	profile	NOUN
cana-5267	78	27	analysis	analysis	NOUN
cana-5267	78	28	to	to	PART
cana-5267	78	29	identify	identify	VERB
cana-5267	78	30	anomalies	anomaly	NOUN
cana-5267	78	31	in	in	ADP
cana-5267	78	32	the	the	DET
cana-5267	78	33	responses	response	NOUN
cana-5267	78	34	produced	produce	VERB
cana-5267	78	35	by	by	ADP
cana-5267	78	36	llms	llm	NOUN
cana-5267	78	37	.	.	PUNCT
cana-5267	79	1	results	result	NOUN
cana-5267	79	2	from	from	ADP
cana-5267	79	3	their	their	PRON
cana-5267	79	4	work	work	NOUN
cana-5267	79	5	proved	prove	VERB
cana-5267	79	6	that	that	SCONJ
cana-5267	79	7	ai	ai	NOUN
cana-5267	79	8	-	-	PUNCT
cana-5267	79	9	based	base	VERB
cana-5267	79	10	systems	system	NOUN
cana-5267	79	11	for	for	ADP
cana-5267	79	12	threat	threat	NOUN
cana-5267	79	13	intelligence	intelligence	NOUN
cana-5267	79	14	make	make	VERB
cana-5267	79	15	it	it	PRON
cana-5267	79	16	nearly	nearly	ADV
cana-5267	79	17	impossible	impossible	ADJ
cana-5267	79	18	to	to	PART
cana-5267	79	19	overlook	overlook	VERB
cana-5267	79	20	cyber	cyber	NOUN
cana-5267	79	21	threats	threat	NOUN
cana-5267	79	22	in	in	ADP
cana-5267	79	23	the	the	DET
cana-5267	79	24	use	use	NOUN
cana-5267	79	25	of	of	ADP
cana-5267	79	26	llms	llm	NOUN
cana-5267	79	27	.	.	PUNCT
cana-5267	80	1	communications	communication	NOUN
cana-5267	80	2	on	on	ADP
cana-5267	80	3	applied	apply	VERB
cana-5267	80	4	nonlinear	nonlinear	ADJ
cana-5267	80	5	analysis	analysis	NOUN
cana-5267	80	6	issn	issn	NOUN
cana-5267	80	7	:	:	PUNCT
cana-5267	80	8	1074	1074	NUM
cana-5267	80	9	-	-	PUNCT
cana-5267	80	10	133x	133x	NUM
cana-5267	80	11	vol	vol	VERB
cana-5267	80	12	32	32	NUM
cana-5267	80	13	no	no	NOUN
cana-5267	80	14	.	.	PUNCT
cana-5267	81	1	10s	10	NOUN
cana-5267	81	2	(	(	PUNCT
cana-5267	81	3	2025	2025	NUM
cana-5267	81	4	)	)	PUNCT
cana-5267	81	5	1662	1662	NUM
cana-5267	81	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5267	81	7	in	in	ADP
cana-5267	81	8	cheong	cheong	PROPN
cana-5267	81	9	's	's	PART
cana-5267	81	10	study	study	NOUN
cana-5267	81	11	,	,	PUNCT
cana-5267	81	12	he	he	PRON
cana-5267	81	13	focused	focus	VERB
cana-5267	81	14	on	on	ADP
cana-5267	81	15	how	how	SCONJ
cana-5267	81	16	hostile	hostile	ADJ
cana-5267	81	17	actions	action	NOUN
cana-5267	81	18	are	be	AUX
cana-5267	81	19	able	able	ADJ
cana-5267	81	20	to	to	PART
cana-5267	81	21	influence	influence	VERB
cana-5267	81	22	llms	llm	NOUN
cana-5267	81	23	which	which	PRON
cana-5267	81	24	includes	include	VERB
cana-5267	81	25	common	common	ADJ
cana-5267	81	26	methods	method	NOUN
cana-5267	81	27	of	of	ADP
cana-5267	81	28	attack	attack	NOUN
cana-5267	81	29	like	like	ADP
cana-5267	81	30	prompt	prompt	ADJ
cana-5267	81	31	injections	injection	NOUN
cana-5267	81	32	,	,	PUNCT
cana-5267	81	33	gradient	gradient	ADJ
cana-5267	81	34	based	base	VERB
cana-5267	81	35	attacks	attack	NOUN
cana-5267	81	36	,	,	PUNCT
cana-5267	81	37	and	and	CCONJ
cana-5267	81	38	evasion	evasion	NOUN
cana-5267	81	39	through	through	ADP
cana-5267	81	40	reinforcement	reinforcement	NOUN
cana-5267	81	41	learning	learning	NOUN
cana-5267	81	42	,	,	PUNCT
cana-5267	81	43	setting	set	VERB
cana-5267	81	44	an	an	DET
cana-5267	81	45	essence	essence	NOUN
cana-5267	81	46	for	for	ADP
cana-5267	81	47	chen	chen	PROPN
cana-5267	81	48	's	's	PART
cana-5267	81	49	extensive	extensive	ADJ
cana-5267	81	50	study	study	NOUN
cana-5267	81	51	.	.	PUNCT
cana-5267	82	1	moreover	moreover	ADV
cana-5267	82	2	,	,	PUNCT
cana-5267	82	3	these	these	DET
cana-5267	82	4	techniques	technique	NOUN
cana-5267	82	5	have	have	AUX
cana-5267	82	6	been	be	AUX
cana-5267	82	7	made	make	VERB
cana-5267	82	8	easier	easy	ADJ
cana-5267	82	9	due	due	ADP
cana-5267	82	10	to	to	ADP
cana-5267	82	11	the	the	DET
cana-5267	82	12	major	major	ADJ
cana-5267	82	13	weak	weak	ADJ
cana-5267	82	14	points	point	NOUN
cana-5267	82	15	left	leave	VERB
cana-5267	82	16	unguarded	unguarded	ADJ
cana-5267	82	17	.	.	PUNCT
cana-5267	83	1	along	along	ADP
cana-5267	83	2	with	with	ADP
cana-5267	83	3	this	this	PRON
cana-5267	83	4	,	,	PUNCT
cana-5267	83	5	more	more	ADJ
cana-5267	83	6	emphasis	emphasis	NOUN
cana-5267	83	7	is	be	AUX
cana-5267	83	8	needed	need	VERB
cana-5267	83	9	towards	towards	ADP
cana-5267	83	10	prone	prone	ADJ
cana-5267	83	11	models	model	NOUN
cana-5267	83	12	as	as	SCONJ
cana-5267	83	13	they	they	PRON
cana-5267	83	14	undergo	undergo	VERB
cana-5267	83	15	heightened	heightened	ADJ
cana-5267	83	16	bullying	bullying	NOUN
cana-5267	83	17	,	,	PUNCT
cana-5267	83	18	so	so	SCONJ
cana-5267	83	19	that	that	SCONJ
cana-5267	83	20	systems	system	NOUN
cana-5267	83	21	are	be	AUX
cana-5267	83	22	not	not	PART
cana-5267	83	23	forced	force	VERB
cana-5267	83	24	to	to	PART
cana-5267	83	25	shut	shut	VERB
cana-5267	83	26	down	down	ADP
cana-5267	83	27	because	because	SCONJ
cana-5267	83	28	of	of	ADP
cana-5267	83	29	unnecessary	unnecessary	ADJ
cana-5267	83	30	obstructions	obstruction	NOUN
cana-5267	83	31	.	.	PUNCT
cana-5267	84	1	the	the	DET
cana-5267	84	2	main	main	ADJ
cana-5267	84	3	point	point	NOUN
cana-5267	84	4	of	of	ADP
cana-5267	84	5	chen	chen	PROPN
cana-5267	84	6	’s	’s	PART
cana-5267	84	7	study	study	NOUN
cana-5267	84	8	was	be	AUX
cana-5267	84	9	the	the	DET
cana-5267	84	10	necessity	necessity	NOUN
cana-5267	84	11	of	of	ADP
cana-5267	84	12	effectuated	effectuated	ADJ
cana-5267	84	13	training	training	NOUN
cana-5267	84	14	which	which	PRON
cana-5267	84	15	would	would	AUX
cana-5267	84	16	strongly	strongly	ADV
cana-5267	84	17	work	work	VERB
cana-5267	84	18	towards	towards	ADP
cana-5267	84	19	building	build	VERB
cana-5267	84	20	higher	high	ADJ
cana-5267	84	21	stamina	stamina	NOUN
cana-5267	84	22	towards	towards	ADP
cana-5267	84	23	aggressive	aggressive	ADJ
cana-5267	84	24	attacks	attack	NOUN
cana-5267	84	25	.	.	PUNCT
cana-5267	85	1	on	on	ADP
cana-5267	85	2	the	the	DET
cana-5267	85	3	other	other	ADJ
cana-5267	85	4	hand	hand	NOUN
cana-5267	85	5	,	,	PUNCT
cana-5267	85	6	chen	chen	PROPN
cana-5267	85	7	focused	focus	VERB
cana-5267	85	8	on	on	ADP
cana-5267	85	9	others	other	NOUN
cana-5267	85	10	such	such	ADJ
cana-5267	85	11	is	be	AUX
cana-5267	85	12	lee	lee	PROPN
cana-5267	85	13	and	and	CCONJ
cana-5267	85	14	park	park	NOUN
cana-5267	85	15	where	where	SCONJ
cana-5267	85	16	they	they	PRON
cana-5267	85	17	focused	focus	VERB
cana-5267	85	18	on	on	ADP
cana-5267	85	19	broader	broad	ADJ
cana-5267	85	20	prompt	prompt	ADJ
cana-5267	85	21	injections	injection	NOUN
cana-5267	85	22	.	.	PUNCT
cana-5267	86	1	these	these	DET
cana-5267	86	2	prompt	prompt	ADJ
cana-5267	86	3	injections	injection	NOUN
cana-5267	86	4	have	have	AUX
cana-5267	86	5	been	be	AUX
cana-5267	86	6	allowed	allow	VERB
cana-5267	86	7	due	due	ADP
cana-5267	86	8	to	to	ADP
cana-5267	86	9	the	the	DET
cana-5267	86	10	lack	lack	NOUN
cana-5267	86	11	of	of	ADP
cana-5267	86	12	effective	effective	ADJ
cana-5267	86	13	means	mean	NOUN
cana-5267	86	14	to	to	PART
cana-5267	86	15	defend	defend	VERB
cana-5267	86	16	resultant	resultant	NOUN
cana-5267	86	17	blanks	blank	NOUN
cana-5267	86	18	made	make	VERB
cana-5267	86	19	intentionally	intentionally	ADV
cana-5267	86	20	by	by	ADP
cana-5267	86	21	adversaries	adversary	NOUN
cana-5267	86	22	which	which	PRON
cana-5267	86	23	wear	wear	VERB
cana-5267	86	24	down	down	ADP
cana-5267	86	25	systems	system	NOUN
cana-5267	86	26	.	.	PUNCT
cana-5267	87	1	these	these	DET
cana-5267	87	2	authors	author	NOUN
cana-5267	87	3	stepped	step	VERB
cana-5267	87	4	in	in	ADV
cana-5267	87	5	to	to	PART
cana-5267	87	6	formulate	formulate	VERB
cana-5267	87	7	a	a	DET
cana-5267	87	8	strategy	strategy	NOUN
cana-5267	87	9	where	where	SCONJ
cana-5267	87	10	llms	llm	NOUN
cana-5267	87	11	can	can	AUX
cana-5267	87	12	automatically	automatically	ADV
cana-5267	87	13	identify	identify	VERB
cana-5267	87	14	and	and	CCONJ
cana-5267	87	15	block	block	VERB
cana-5267	87	16	hijacking	hijacking	NOUN
cana-5267	87	17	prompts	prompt	NOUN
cana-5267	87	18	which	which	PRON
cana-5267	87	19	help	help	VERB
cana-5267	87	20	dynamic	dynamic	ADJ
cana-5267	87	21	defending	defending	NOUN
cana-5267	87	22	encourages	encourage	VERB
cana-5267	87	23	discipline	discipline	NOUN
cana-5267	87	24	.	.	PUNCT
cana-5267	88	1	as	as	ADP
cana-5267	88	2	a	a	DET
cana-5267	88	3	result	result	NOUN
cana-5267	88	4	,	,	PUNCT
cana-5267	88	5	there	there	PRON
cana-5267	88	6	is	be	VERB
cana-5267	88	7	clear	clear	ADJ
cana-5267	88	8	sight	sight	NOUN
cana-5267	88	9	in	in	ADP
cana-5267	88	10	the	the	DET
cana-5267	88	11	chances	chance	NOUN
cana-5267	88	12	of	of	ADP
cana-5267	88	13	being	be	AUX
cana-5267	88	14	right	right	ADJ
cana-5267	88	15	and	and	CCONJ
cana-5267	88	16	rid	rid	VERB
cana-5267	88	17	of	of	ADP
cana-5267	88	18	constructive	constructive	ADJ
cana-5267	88	19	blocking	blocking	NOUN
cana-5267	88	20	or	or	CCONJ
cana-5267	88	21	deception	deception	NOUN
cana-5267	88	22	.	.	PUNCT
cana-5267	89	1	the	the	DET
cana-5267	89	2	idea	idea	NOUN
cana-5267	89	3	of	of	ADP
cana-5267	89	4	model	model	NOUN
cana-5267	89	5	inversion	inversion	NOUN
cana-5267	89	6	attacks	attack	NOUN
cana-5267	89	7	has	have	AUX
cana-5267	89	8	become	become	VERB
cana-5267	89	9	more	more	ADV
cana-5267	89	10	popular	popular	ADJ
cana-5267	89	11	as	as	SCONJ
cana-5267	89	12	more	more	ADJ
cana-5267	89	13	people	people	NOUN
cana-5267	89	14	try	try	VERB
cana-5267	89	15	to	to	PART
cana-5267	89	16	obtain	obtain	VERB
cana-5267	89	17	sensitive	sensitive	ADJ
cana-5267	89	18	data	datum	NOUN
cana-5267	89	19	from	from	ADP
cana-5267	89	20	llms	llm	NOUN
cana-5267	89	21	by	by	ADP
cana-5267	89	22	deconstructing	deconstruct	VERB
cana-5267	89	23	the	the	DET
cana-5267	89	24	components	component	NOUN
cana-5267	89	25	of	of	ADP
cana-5267	89	26	their	their	PRON
cana-5267	89	27	training	training	NOUN
cana-5267	89	28	data	datum	NOUN
cana-5267	89	29	.	.	PUNCT
cana-5267	90	1	wang	wang	PROPN
cana-5267	90	2	and	and	CCONJ
cana-5267	90	3	xu	xu	PROPN
cana-5267	91	1	[	[	X
cana-5267	91	2	7	7	X
cana-5267	91	3	]	]	PUNCT
cana-5267	91	4	studied	study	VERB
cana-5267	91	5	the	the	DET
cana-5267	91	6	problem	problem	NOUN
cana-5267	91	7	of	of	ADP
cana-5267	91	8	model	model	NOUN
cana-5267	91	9	inversion	inversion	NOUN
cana-5267	91	10	on	on	ADP
cana-5267	91	11	transformer	transformer	NOUN
cana-5267	91	12	architectures	architecture	NOUN
cana-5267	91	13	and	and	CCONJ
cana-5267	91	14	demonstrated	demonstrate	VERB
cana-5267	91	15	that	that	SCONJ
cana-5267	91	16	even	even	ADV
cana-5267	91	17	model	model	NOUN
cana-5267	91	18	outputs	output	NOUN
cana-5267	91	19	can	can	AUX
cana-5267	91	20	be	be	AUX
cana-5267	91	21	exploited	exploit	VERB
cana-5267	91	22	through	through	ADP
cana-5267	91	23	adversarial	adversarial	ADJ
cana-5267	91	24	querying	query	VERB
cana-5267	91	25	to	to	PART
cana-5267	91	26	infer	infer	VERB
cana-5267	91	27	private	private	ADJ
cana-5267	91	28	data	datum	NOUN
cana-5267	91	29	.	.	PUNCT
cana-5267	92	1	their	their	PRON
cana-5267	92	2	results	result	NOUN
cana-5267	92	3	highlight	highlight	VERB
cana-5267	92	4	the	the	DET
cana-5267	92	5	requirement	requirement	NOUN
cana-5267	92	6	of	of	ADP
cana-5267	92	7	implementing	implement	VERB
cana-5267	92	8	differential	differential	ADJ
cana-5267	92	9	privacy	privacy	NOUN
cana-5267	92	10	to	to	PART
cana-5267	92	11	avoid	avoid	VERB
cana-5267	92	12	unintentional	unintentional	ADJ
cana-5267	92	13	data	datum	NOUN
cana-5267	92	14	leakage	leakage	NOUN
cana-5267	92	15	in	in	ADP
cana-5267	92	16	ai	ai	PROPN
cana-5267	92	17	applications	application	NOUN
cana-5267	92	18	.	.	PUNCT
cana-5267	93	1	patel	patel	NOUN
cana-5267	93	2	and	and	CCONJ
cana-5267	93	3	shah	shah	PROPN
cana-5267	93	4	[	[	X
cana-5267	93	5	8	8	NUM
cana-5267	93	6	]	]	PUNCT
cana-5267	93	7	also	also	ADV
cana-5267	93	8	researched	research	VERB
cana-5267	93	9	ways	way	NOUN
cana-5267	93	10	of	of	ADP
cana-5267	93	11	homomorphic	homomorphic	ADJ
cana-5267	93	12	encryption	encryption	NOUN
cana-5267	93	13	and	and	CCONJ
cana-5267	93	14	federated	federated	ADJ
cana-5267	93	15	learning	learning	NOUN
cana-5267	93	16	to	to	PART
cana-5267	93	17	defend	defend	VERB
cana-5267	93	18	llm	llm	PROPN
cana-5267	93	19	implementations	implementation	NOUN
cana-5267	93	20	without	without	ADP
cana-5267	93	21	undermining	undermine	VERB
cana-5267	93	22	the	the	DET
cana-5267	93	23	model	model	NOUN
cana-5267	93	24	’s	’s	PART
cana-5267	93	25	efficacy	efficacy	NOUN
cana-5267	93	26	.	.	PUNCT
cana-5267	94	1	their	their	PRON
cana-5267	94	2	study	study	NOUN
cana-5267	94	3	asserts	assert	VERB
cana-5267	94	4	that	that	SCONJ
cana-5267	94	5	privacy	privacy	NOUN
cana-5267	94	6	-	-	PUNCT
cana-5267	94	7	preserving	preserve	VERB
cana-5267	94	8	ai	ai	NOUN
cana-5267	94	9	methodologies	methodology	NOUN
cana-5267	94	10	like	like	ADP
cana-5267	94	11	secret	secret	ADJ
cana-5267	94	12	multi	multi	ADJ
cana-5267	94	13	-	-	ADJ
cana-5267	94	14	party	party	ADJ
cana-5267	94	15	computation	computation	NOUN
cana-5267	94	16	(	(	PUNCT
cana-5267	94	17	smpc	smpc	NOUN
cana-5267	94	18	)	)	PUNCT
cana-5267	94	19	are	be	AUX
cana-5267	94	20	a	a	DET
cana-5267	94	21	suitable	suitable	ADJ
cana-5267	94	22	approach	approach	NOUN
cana-5267	94	23	to	to	PART
cana-5267	94	24	block	block	VERB
cana-5267	94	25	the	the	DET
cana-5267	94	26	exploitation	exploitation	NOUN
cana-5267	94	27	of	of	ADP
cana-5267	94	28	confidential	confidential	ADJ
cana-5267	94	29	information	information	NOUN
cana-5267	94	30	captured	capture	VERB
cana-5267	94	31	in	in	ADP
cana-5267	94	32	the	the	DET
cana-5267	94	33	model	model	NOUN
cana-5267	94	34	.	.	PUNCT
cana-5267	95	1	outside	outside	ADP
cana-5267	95	2	of	of	ADP
cana-5267	95	3	the	the	DET
cana-5267	95	4	more	more	ADV
cana-5267	95	5	focused	focused	ADJ
cana-5267	95	6	attack	attack	NOUN
cana-5267	95	7	vectors	vector	NOUN
cana-5267	95	8	,	,	PUNCT
cana-5267	95	9	researchers	researcher	NOUN
cana-5267	95	10	have	have	AUX
cana-5267	95	11	looked	look	VERB
cana-5267	95	12	into	into	ADP
cana-5267	95	13	holistic	holistic	ADJ
cana-5267	95	14	comprehensive	comprehensive	ADJ
cana-5267	95	15	frameworks	framework	NOUN
cana-5267	95	16	for	for	ADP
cana-5267	95	17	risk	risk	NOUN
cana-5267	95	18	assessment	assessment	NOUN
cana-5267	95	19	in	in	ADP
cana-5267	95	20	the	the	DET
cana-5267	95	21	context	context	NOUN
cana-5267	95	22	of	of	ADP
cana-5267	95	23	llm	llm	PROPN
cana-5267	95	24	systems	system	NOUN
cana-5267	95	25	integration	integration	NOUN
cana-5267	95	26	.	.	PUNCT
cana-5267	96	1	hall	hall	NOUN
cana-5267	96	2	and	and	CCONJ
cana-5267	96	3	peterson	peterson	PROPN
cana-5267	96	4	[	[	X
cana-5267	96	5	9	9	NUM
cana-5267	96	6	]	]	PUNCT
cana-5267	96	7	created	create	VERB
cana-5267	96	8	a	a	DET
cana-5267	96	9	model	model	NOUN
cana-5267	96	10	of	of	ADP
cana-5267	96	11	threat	threat	NOUN
cana-5267	96	12	intelligence	intelligence	NOUN
cana-5267	96	13	-	-	PUNCT
cana-5267	96	14	driven	drive	VERB
cana-5267	96	15	risk	risk	NOUN
cana-5267	96	16	assessment	assessment	NOUN
cana-5267	96	17	that	that	PRON
cana-5267	96	18	is	be	AUX
cana-5267	96	19	based	base	VERB
cana-5267	96	20	on	on	ADP
cana-5267	96	21	continuous	continuous	ADJ
cana-5267	96	22	monitoring	monitoring	NOUN
cana-5267	96	23	of	of	ADP
cana-5267	96	24	ai	ai	PROPN
cana-5267	96	25	implementations	implementation	NOUN
cana-5267	96	26	for	for	ADP
cana-5267	96	27	dynamic	dynamic	ADJ
cana-5267	96	28	security	security	NOUN
cana-5267	96	29	risks	risk	NOUN
cana-5267	96	30	.	.	PUNCT
cana-5267	97	1	this	this	DET
cana-5267	97	2	model	model	NOUN
cana-5267	97	3	associates	associates	PROPN
cana-5267	97	4	cti	cti	PROPN
cana-5267	97	5	feeds	feed	VERB
cana-5267	97	6	with	with	ADP
cana-5267	97	7	automated	automate	VERB
cana-5267	97	8	anomaly	anomaly	NOUN
cana-5267	97	9	detection	detection	NOUN
cana-5267	97	10	systems	system	NOUN
cana-5267	97	11	for	for	ADP
cana-5267	97	12	real	real	ADJ
cana-5267	97	13	-	-	PUNCT
cana-5267	97	14	time	time	NOUN
cana-5267	97	15	mitigation	mitigation	NOUN
cana-5267	97	16	.	.	PUNCT
cana-5267	98	1	nguyen	nguyen	NOUN
cana-5267	98	2	and	and	CCONJ
cana-5267	98	3	bui	bui	NOUN
cana-5267	99	1	[	[	X
cana-5267	99	2	10	10	NUM
cana-5267	99	3	]	]	PUNCT
cana-5267	99	4	adapted	adapt	VERB
cana-5267	99	5	this	this	DET
cana-5267	99	6	model	model	NOUN
cana-5267	99	7	to	to	PART
cana-5267	99	8	include	include	VERB
cana-5267	99	9	deep	deep	ADJ
cana-5267	99	10	reinforcement	reinforcement	NOUN
cana-5267	99	11	learning	learning	NOUN
cana-5267	99	12	(	(	PUNCT
cana-5267	99	13	drl	drl	PROPN
cana-5267	99	14	)	)	PUNCT
cana-5267	99	15	as	as	ADP
cana-5267	99	16	a	a	DET
cana-5267	99	17	component	component	NOUN
cana-5267	99	18	of	of	ADP
cana-5267	99	19	cyber	cyber	ADJ
cana-5267	99	20	risk	risk	NOUN
cana-5267	99	21	management	management	NOUN
cana-5267	99	22	.	.	PUNCT
cana-5267	100	1	their	their	PRON
cana-5267	100	2	findings	finding	NOUN
cana-5267	100	3	indicate	indicate	VERB
cana-5267	100	4	that	that	SCONJ
cana-5267	100	5	adaptive	adaptive	ADJ
cana-5267	100	6	learning	learning	NOUN
cana-5267	100	7	-	-	PUNCT
cana-5267	100	8	enabled	enable	VERB
cana-5267	100	9	security	security	NOUN
cana-5267	100	10	frameworks	framework	NOUN
cana-5267	100	11	can	can	AUX
cana-5267	100	12	identify	identify	VERB
cana-5267	100	13	new	new	ADJ
cana-5267	100	14	attack	attack	NOUN
cana-5267	100	15	patterns	pattern	NOUN
cana-5267	100	16	and	and	CCONJ
cana-5267	100	17	streamline	streamline	VERB
cana-5267	100	18	defense	defense	NOUN
cana-5267	100	19	processes	process	NOUN
cana-5267	100	20	in	in	ADP
cana-5267	100	21	llm	llm	PROPN
cana-5267	100	22	environments	environment	NOUN
cana-5267	100	23	.	.	PUNCT
cana-5267	101	1	another	another	DET
cana-5267	101	2	equally	equally	ADV
cana-5267	101	3	compelling	compelling	ADJ
cana-5267	101	4	topic	topic	NOUN
cana-5267	101	5	is	be	AUX
cana-5267	101	6	security	security	NOUN
cana-5267	101	7	monitoring	monitoring	NOUN
cana-5267	101	8	in	in	ADP
cana-5267	101	9	llm	llm	ADV
cana-5267	101	10	-	-	PUNCT
cana-5267	101	11	driven	drive	VERB
cana-5267	101	12	applications	application	NOUN
cana-5267	101	13	.	.	PUNCT
cana-5267	102	1	qureshi	qureshi	PROPN
cana-5267	102	2	and	and	CCONJ
cana-5267	102	3	khan	khan	PROPN
cana-5267	103	1	[	[	X
cana-5267	103	2	11	11	NUM
cana-5267	103	3	]	]	PUNCT
cana-5267	103	4	designed	design	VERB
cana-5267	103	5	an	an	DET
cana-5267	103	6	intrusion	intrusion	NOUN
cana-5267	103	7	detection	detection	NOUN
cana-5267	103	8	system	system	NOUN
cana-5267	103	9	(	(	PUNCT
cana-5267	103	10	ids	id	NOUN
cana-5267	103	11	)	)	PUNCT
cana-5267	103	12	for	for	ADP
cana-5267	103	13	ai	ai	ADV
cana-5267	103	14	-	-	PUNCT
cana-5267	103	15	based	base	VERB
cana-5267	103	16	chatbots	chatbot	NOUN
cana-5267	103	17	that	that	PRON
cana-5267	103	18	works	work	VERB
cana-5267	103	19	in	in	ADP
cana-5267	103	20	realtime	realtime	NOUN
cana-5267	103	21	.	.	PUNCT
cana-5267	104	1	their	their	PRON
cana-5267	104	2	work	work	NOUN
cana-5267	104	3	reveals	reveal	VERB
cana-5267	104	4	that	that	SCONJ
cana-5267	104	5	hybrid	hybrid	NOUN
cana-5267	104	6	ml	ml	NOUN
cana-5267	104	7	models	model	NOUN
cana-5267	104	8	that	that	PRON
cana-5267	104	9	incorporate	incorporate	VERB
cana-5267	104	10	nlp	nlp	ADJ
cana-5267	104	11	-	-	PUNCT
cana-5267	104	12	driven	drive	VERB
cana-5267	104	13	sentiment	sentiment	NOUN
cana-5267	104	14	analysis	analysis	NOUN
cana-5267	104	15	and	and	CCONJ
cana-5267	104	16	anomauy	anomauy	NOUN
cana-5267	104	17	detection	detection	NOUN
cana-5267	104	18	provide	provide	VERB
cana-5267	104	19	effective	effective	ADJ
cana-5267	104	20	real	real	ADJ
cana-5267	104	21	-	-	PUNCT
cana-5267	104	22	time	time	NOUN
cana-5267	104	23	recognition	recognition	NOUN
cana-5267	104	24	of	of	ADP
cana-5267	104	25	possible	possible	ADJ
cana-5267	104	26	security	security	NOUN
cana-5267	104	27	threats	threat	NOUN
cana-5267	104	28	during	during	ADP
cana-5267	104	29	user	user	NOUN
cana-5267	104	30	interactions	interaction	NOUN
cana-5267	104	31	.	.	PUNCT
cana-5267	105	1	in	in	ADP
cana-5267	105	2	the	the	DET
cana-5267	105	3	same	same	ADJ
cana-5267	105	4	manner	manner	NOUN
cana-5267	105	5	,	,	PUNCT
cana-5267	105	6	raj	raj	PROPN
cana-5267	105	7	and	and	CCONJ
cana-5267	105	8	kumar	kumar	PROPN
cana-5267	105	9	[	[	X
cana-5267	105	10	12	12	NUM
cana-5267	105	11	]	]	PUNCT
cana-5267	105	12	explored	explore	VERB
cana-5267	105	13	attack	attack	NOUN
cana-5267	105	14	defenses	defense	NOUN
cana-5267	105	15	against	against	ADP
cana-5267	105	16	model	model	NOUN
cana-5267	105	17	inversion	inversion	NOUN
cana-5267	105	18	and	and	CCONJ
cana-5267	105	19	other	other	ADJ
cana-5267	105	20	adversarial	adversarial	ADJ
cana-5267	105	21	attacks	attack	NOUN
cana-5267	105	22	,	,	PUNCT
cana-5267	105	23	proposing	propose	VERB
cana-5267	105	24	the	the	DET
cana-5267	105	25	use	use	NOUN
cana-5267	105	26	of	of	ADP
cana-5267	105	27	explainable	explainable	ADJ
cana-5267	105	28	ai	ai	NOUN
cana-5267	105	29	(	(	PUNCT
cana-5267	105	30	xai	xai	PROPN
cana-5267	105	31	)	)	PUNCT
cana-5267	105	32	to	to	PART
cana-5267	105	33	enhance	enhance	VERB
cana-5267	105	34	ai	ai	VERB
cana-5267	105	35	transparency	transparency	NOUN
cana-5267	105	36	about	about	ADP
cana-5267	105	37	the	the	DET
cana-5267	105	38	security	security	NOUN
cana-5267	105	39	decisions	decision	NOUN
cana-5267	105	40	that	that	PRON
cana-5267	105	41	were	be	AUX
cana-5267	105	42	executed	execute	VERB
cana-5267	105	43	.	.	PUNCT
cana-5267	106	1	their	their	PRON
cana-5267	106	2	research	research	NOUN
cana-5267	106	3	argues	argue	VERB
cana-5267	106	4	that	that	SCONJ
cana-5267	106	5	such	such	ADJ
cana-5267	106	6	interpretable	interpretable	ADJ
cana-5267	106	7	threat	threat	NOUN
cana-5267	106	8	detection	detection	NOUN
cana-5267	106	9	frameworks	framework	NOUN
cana-5267	106	10	can	can	AUX
cana-5267	106	11	foster	foster	VERB
cana-5267	106	12	user	user	NOUN
cana-5267	106	13	confidence	confidence	NOUN
cana-5267	106	14	in	in	ADP
cana-5267	106	15	ai	ai	ADJ
cana-5267	106	16	-	-	PUNCT
cana-5267	106	17	powered	power	VERB
cana-5267	106	18	cybersecurity	cybersecurity	NOUN
cana-5267	106	19	systems	system	NOUN
cana-5267	106	20	.	.	PUNCT
cana-5267	107	1	the	the	DET
cana-5267	107	2	application	application	NOUN
cana-5267	107	3	of	of	ADP
cana-5267	107	4	automated	automate	VERB
cana-5267	107	5	threat	threat	NOUN
cana-5267	107	6	modeling	modeling	NOUN
cana-5267	107	7	to	to	ADP
cana-5267	107	8	nlp	nlp	NOUN
cana-5267	107	9	systems	system	NOUN
cana-5267	107	10	requires	require	VERB
cana-5267	107	11	improving	improve	VERB
cana-5267	107	12	dataset	dataset	ADJ
cana-5267	107	13	accuracy	accuracy	NOUN
cana-5267	107	14	and	and	CCONJ
cana-5267	107	15	building	building	NOUN
cana-5267	107	16	secure	secure	ADJ
cana-5267	107	17	ai	ai	VERB
cana-5267	107	18	pipelines	pipeline	NOUN
cana-5267	107	19	.	.	PUNCT
cana-5267	108	1	in	in	ADP
cana-5267	108	2	the	the	DET
cana-5267	108	3	realm	realm	NOUN
cana-5267	108	4	of	of	ADP
cana-5267	108	5	cybersecurity	cybersecurity	NOUN
cana-5267	108	6	,	,	PUNCT
cana-5267	108	7	smith	smith	PROPN
cana-5267	108	8	and	and	CCONJ
cana-5267	108	9	taylor	taylor	PROPN
cana-5267	108	10	[	[	X
cana-5267	108	11	13	13	NUM
cana-5267	108	12	]	]	PUNCT
cana-5267	108	13	studied	study	VERB
cana-5267	108	14	the	the	DET
cana-5267	108	15	communications	communication	NOUN
cana-5267	108	16	on	on	ADP
cana-5267	108	17	applied	apply	VERB
cana-5267	108	18	nonlinear	nonlinear	ADJ
cana-5267	108	19	analysis	analysis	NOUN
cana-5267	108	20	issn	issn	NOUN
cana-5267	108	21	:	:	PUNCT
cana-5267	108	22	1074	1074	NUM
cana-5267	108	23	-	-	PUNCT
cana-5267	108	24	133x	133x	NUM
cana-5267	108	25	vol	vol	VERB
cana-5267	108	26	32	32	NUM
cana-5267	108	27	no	no	NOUN
cana-5267	108	28	.	.	PUNCT
cana-5267	109	1	10s	10	NOUN
cana-5267	109	2	(	(	PUNCT
cana-5267	109	3	2025	2025	NUM
cana-5267	109	4	)	)	PUNCT
cana-5267	109	5	1663	1663	NUM
cana-5267	109	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5267	109	7	application	application	NOUN
cana-5267	109	8	of	of	ADP
cana-5267	109	9	nlp	nlp	NOUN
cana-5267	109	10	technologies	technology	NOUN
cana-5267	110	1	and	and	CCONJ
cana-5267	111	1	demonstrated	demonstrate	VERB
cana-5267	111	2	that	that	SCONJ
cana-5267	111	3	llm	llm	NOUN
cana-5267	111	4	-	-	PUNCT
cana-5267	111	5	based	base	VERB
cana-5267	111	6	anomaly	anomaly	NOUN
cana-5267	111	7	detection	detection	NOUN
cana-5267	111	8	systems	system	NOUN
cana-5267	111	9	outperform	outperform	VERB
cana-5267	111	10	traditional	traditional	ADJ
cana-5267	111	11	rule	rule	NOUN
cana-5267	111	12	-	-	PUNCT
cana-5267	111	13	based	base	VERB
cana-5267	111	14	security	security	NOUN
cana-5267	111	15	log	log	NOUN
cana-5267	111	16	analysis	analysis	NOUN
cana-5267	111	17	systems	system	NOUN
cana-5267	111	18	in	in	ADP
cana-5267	111	19	scaling	scaling	NOUN
cana-5267	111	20	and	and	CCONJ
cana-5267	111	21	efficiency	efficiency	NOUN
cana-5267	111	22	processing	process	VERB
cana-5267	111	23	massive	massive	ADJ
cana-5267	111	24	security	security	NOUN
cana-5267	111	25	logs	log	NOUN
cana-5267	111	26	.	.	PUNCT
cana-5267	112	1	it	it	PRON
cana-5267	112	2	was	be	AUX
cana-5267	112	3	determined	determined	ADJ
cana-5267	112	4	that	that	SCONJ
cana-5267	112	5	ai	ai	VERB
cana-5267	112	6	-	-	PUNCT
cana-5267	112	7	enabled	enable	VERB
cana-5267	112	8	nlp	nlp	NOUN
cana-5267	112	9	models	model	NOUN
cana-5267	112	10	can	can	AUX
cana-5267	112	11	formulate	formulate	VERB
cana-5267	112	12	relevant	relevant	ADJ
cana-5267	112	13	responses	response	NOUN
cana-5267	112	14	in	in	ADP
cana-5267	112	15	dealing	deal	VERB
cana-5267	112	16	with	with	ADP
cana-5267	112	17	unstructured	unstructured	ADJ
cana-5267	112	18	security	security	NOUN
cana-5267	112	19	data	datum	NOUN
cana-5267	112	20	,	,	PUNCT
cana-5267	112	21	thus	thus	ADV
cana-5267	112	22	enhancing	enhance	VERB
cana-5267	112	23	the	the	DET
cana-5267	112	24	automated	automate	VERB
cana-5267	112	25	threat	threat	NOUN
cana-5267	112	26	evaluation	evaluation	NOUN
cana-5267	112	27	processes	process	NOUN
cana-5267	112	28	.	.	PUNCT
cana-5267	113	1	zhou	zhou	PROPN
cana-5267	113	2	and	and	CCONJ
cana-5267	113	3	lin	lin	PROPN
cana-5267	114	1	[	[	X
cana-5267	114	2	14	14	NUM
cana-5267	114	3	]	]	PUNCT
cana-5267	114	4	delved	delve	VERB
cana-5267	114	5	into	into	ADP
cana-5267	114	6	machine	machine	NOUN
cana-5267	114	7	learning	learn	VERB
cana-5267	114	8	vulnerability	vulnerability	NOUN
cana-5267	114	9	detection	detection	NOUN
cana-5267	114	10	frameworks	framework	NOUN
cana-5267	114	11	,	,	PUNCT
cana-5267	114	12	applying	apply	VERB
cana-5267	114	13	unsupervised	unsupervised	ADJ
cana-5267	114	14	anomaly	anomaly	NOUN
cana-5267	114	15	detection	detection	NOUN
cana-5267	114	16	methods	method	NOUN
cana-5267	114	17	to	to	PART
cana-5267	114	18	expose	expose	VERB
cana-5267	114	19	vulnerabilities	vulnerability	NOUN
cana-5267	114	20	within	within	ADP
cana-5267	114	21	ai	ai	ADJ
cana-5267	114	22	deployment	deployment	NOUN
cana-5267	114	23	pipelines	pipeline	NOUN
cana-5267	114	24	.	.	PUNCT
cana-5267	115	1	their	their	PRON
cana-5267	115	2	findings	finding	NOUN
cana-5267	115	3	advocate	advocate	VERB
cana-5267	115	4	the	the	DET
cana-5267	115	5	position	position	NOUN
cana-5267	115	6	that	that	PRON
cana-5267	115	7	audits	audits	NOUN
cana-5267	115	8	on	on	ADP
cana-5267	115	9	security	security	NOUN
cana-5267	115	10	made	make	VERB
cana-5267	115	11	by	by	ADP
cana-5267	115	12	ai	ai	NOUN
cana-5267	115	13	can	can	AUX
cana-5267	115	14	improve	improve	VERB
cana-5267	115	15	compliance	compliance	NOUN
cana-5267	115	16	and	and	CCONJ
cana-5267	115	17	control	control	NOUN
cana-5267	115	18	for	for	ADP
cana-5267	115	19	regulated	regulated	ADJ
cana-5267	115	20	standards	standard	NOUN
cana-5267	115	21	within	within	ADP
cana-5267	115	22	llm	llm	NOUN
cana-5267	115	23	-	-	PUNCT
cana-5267	115	24	enabled	enable	VERB
cana-5267	115	25	systems	system	NOUN
cana-5267	115	26	.	.	PUNCT
cana-5267	116	1	finally	finally	ADV
cana-5267	116	2	,	,	PUNCT
cana-5267	116	3	wang	wang	PROPN
cana-5267	116	4	et	et	PROPN
cana-5267	116	5	al	al	PROPN
cana-5267	116	6	.	.	PUNCT
cana-5267	117	1	[	[	X
cana-5267	117	2	15	15	NUM
cana-5267	117	3	]	]	PUNCT
cana-5267	117	4	created	create	VERB
cana-5267	117	5	an	an	DET
cana-5267	117	6	ai	ai	NOUN
cana-5267	117	7	security	security	NOUN
cana-5267	117	8	model	model	NOUN
cana-5267	117	9	that	that	PRON
cana-5267	117	10	integrates	integrate	VERB
cana-5267	117	11	machine	machine	NOUN
cana-5267	117	12	learning	learn	VERB
cana-5267	117	13	with	with	ADP
cana-5267	117	14	cybersecurity	cybersecurity	NOUN
cana-5267	117	15	analytics	analytic	NOUN
cana-5267	117	16	and	and	CCONJ
cana-5267	117	17	behavioral	behavioral	ADJ
cana-5267	117	18	profiling	profiling	NOUN
cana-5267	117	19	to	to	ADP
cana-5267	117	20	automate	automate	ADJ
cana-5267	117	21	threat	threat	NOUN
cana-5267	117	22	mitigation	mitigation	NOUN
cana-5267	117	23	and	and	CCONJ
cana-5267	117	24	detection	detection	NOUN
cana-5267	117	25	strategies	strategy	NOUN
cana-5267	117	26	for	for	ADP
cana-5267	117	27	llms	llm	NOUN
cana-5267	117	28	.	.	PUNCT
cana-5267	118	1	their	their	PRON
cana-5267	118	2	proposal	proposal	NOUN
cana-5267	118	3	employs	employ	VERB
cana-5267	118	4	hybrid	hybrid	ADJ
cana-5267	118	5	ai	ai	NOUN
cana-5267	118	6	techniques	technique	NOUN
cana-5267	118	7	,	,	PUNCT
cana-5267	118	8	including	include	VERB
cana-5267	118	9	self	self	NOUN
cana-5267	118	10	-	-	PUNCT
cana-5267	118	11	learning	learn	VERB
cana-5267	118	12	anomaly	anomaly	NOUN
cana-5267	118	13	detection	detection	NOUN
cana-5267	118	14	systems	system	NOUN
cana-5267	118	15	and	and	CCONJ
cana-5267	118	16	nlp	nlp	NOUN
cana-5267	118	17	-	-	PUNCT
cana-5267	118	18	based	base	VERB
cana-5267	118	19	threat	threat	NOUN
cana-5267	118	20	intelligence	intelligence	NOUN
cana-5267	118	21	,	,	PUNCT
cana-5267	118	22	enhancing	enhance	VERB
cana-5267	118	23	the	the	DET
cana-5267	118	24	security	security	NOUN
cana-5267	118	25	posture	posture	NOUN
cana-5267	118	26	of	of	ADP
cana-5267	118	27	enterprise	enterprise	NOUN
cana-5267	118	28	-	-	PUNCT
cana-5267	118	29	level	level	NOUN
cana-5267	118	30	llm	llm	NOUN
cana-5267	118	31	systems	system	NOUN
cana-5267	118	32	.	.	PUNCT
cana-5267	119	1	research	research	NOUN
cana-5267	119	2	methodology	methodology	NOUN
cana-5267	119	3	data	datum	NOUN
cana-5267	119	4	sources	source	NOUN
cana-5267	119	5	layer	layer	NOUN
cana-5267	119	6	:	:	PUNCT
cana-5267	119	7	this	this	DET
cana-5267	119	8	layer	layer	NOUN
cana-5267	119	9	focuses	focus	VERB
cana-5267	119	10	on	on	ADP
cana-5267	119	11	threat	threat	NOUN
cana-5267	119	12	collection	collection	NOUN
cana-5267	119	13	and	and	CCONJ
cana-5267	119	14	mitigation	mitigation	NOUN
cana-5267	119	15	and	and	CCONJ
cana-5267	119	16	is	be	AUX
cana-5267	119	17	the	the	DET
cana-5267	119	18	foundation	foundation	NOUN
cana-5267	119	19	for	for	ADP
cana-5267	119	20	the	the	DET
cana-5267	119	21	automated	automate	VERB
cana-5267	119	22	threat	threat	NOUN
cana-5267	119	23	modeling	model	VERB
cana-5267	119	24	system	system	NOUN
cana-5267	119	25	based	base	VERB
cana-5267	119	26	on	on	ADP
cana-5267	119	27	llm	llm	PROPN
cana-5267	119	28	integration	integration	NOUN
cana-5267	119	29	.	.	PUNCT
cana-5267	120	1	it	it	PRON
cana-5267	120	2	collects	collect	VERB
cana-5267	120	3	security	security	NOUN
cana-5267	120	4	logs	log	NOUN
cana-5267	120	5	,	,	PUNCT
cana-5267	120	6	adversarial	adversarial	ADJ
cana-5267	120	7	attack	attack	NOUN
cana-5267	120	8	datasets	dataset	NOUN
cana-5267	120	9	,	,	PUNCT
cana-5267	120	10	system	system	NOUN
cana-5267	120	11	activity	activity	NOUN
cana-5267	120	12	reports	report	NOUN
cana-5267	120	13	,	,	PUNCT
cana-5267	120	14	cve	cve	NOUN
cana-5267	120	15	and	and	CCONJ
cana-5267	120	16	mitre	mitre	NOUN
cana-5267	120	17	vulnerabilities	vulnerability	NOUN
cana-5267	120	18	,	,	PUNCT
cana-5267	120	19	and	and	CCONJ
cana-5267	120	20	live	live	ADJ
cana-5267	120	21	network	network	NOUN
cana-5267	120	22	traffic	traffic	NOUN
cana-5267	120	23	data	datum	NOUN
cana-5267	120	24	.	.	PUNCT
cana-5267	121	1	both	both	PRON
cana-5267	121	2	structured	structure	VERB
cana-5267	121	3	and	and	CCONJ
cana-5267	121	4	unstructured	unstructured	ADJ
cana-5267	121	5	data	datum	NOUN
cana-5267	121	6	are	be	AUX
cana-5267	121	7	collected	collect	VERB
cana-5267	121	8	through	through	ADP
cana-5267	121	9	apis	apis	ADJ
cana-5267	121	10	,	,	PUNCT
cana-5267	121	11	ullm	ullm	ADJ
cana-5267	121	12	authentication	authentication	NOUN
cana-5267	121	13	attempts	attempt	VERB
cana-5267	121	14	,	,	PUNCT
cana-5267	121	15	and	and	CCONJ
cana-5267	121	16	anomaly	anomaly	NOUN
cana-5267	121	17	detection	detection	NOUN
cana-5267	121	18	systems	system	NOUN
cana-5267	121	19	,	,	PUNCT
cana-5267	121	20	which	which	PRON
cana-5267	121	21	guarantee	guarantee	VERB
cana-5267	121	22	comprehensive	comprehensive	ADJ
cana-5267	121	23	threat	threat	NOUN
cana-5267	121	24	intelligence	intelligence	NOUN
cana-5267	121	25	.	.	PUNCT
cana-5267	122	1	detection	detection	NOUN
cana-5267	122	2	of	of	ADP
cana-5267	122	3	suspicious	suspicious	ADJ
cana-5267	122	4	activities	activity	NOUN
cana-5267	122	5	encompasses	encompass	VERB
cana-5267	122	6	unauthorized	unauthorized	ADJ
cana-5267	122	7	access	access	NOUN
cana-5267	122	8	along	along	ADP
cana-5267	122	9	with	with	ADP
cana-5267	122	10	adversarial	adversarial	ADJ
cana-5267	122	11	attempts	attempt	NOUN
cana-5267	122	12	and	and	CCONJ
cana-5267	122	13	patterns	pattern	NOUN
cana-5267	122	14	employing	employ	VERB
cana-5267	122	15	the	the	DET
cana-5267	122	16	trained	train	VERB
cana-5267	122	17	ml	ml	NOUN
cana-5267	122	18	-	-	PUNCT
cana-5267	122	19	based	base	VERB
cana-5267	122	20	threat	threat	NOUN
cana-5267	122	21	detection	detection	NOUN
cana-5267	122	22	model	model	NOUN
cana-5267	122	23	with	with	ADP
cana-5267	122	24	historical	historical	ADJ
cana-5267	122	25	and	and	CCONJ
cana-5267	122	26	real	real	ADJ
cana-5267	122	27	-	-	PUNCT
cana-5267	122	28	time	time	NOUN
cana-5267	122	29	data	datum	NOUN
cana-5267	122	30	.	.	PUNCT
cana-5267	123	1	moreover	moreover	ADV
cana-5267	123	2	,	,	PUNCT
cana-5267	123	3	it	it	PRON
cana-5267	123	4	consists	consist	VERB
cana-5267	123	5	of	of	ADP
cana-5267	123	6	user	user	NOUN
cana-5267	123	7	behavior	behavior	NOUN
cana-5267	123	8	analytics	analytic	NOUN
cana-5267	123	9	(	(	PUNCT
cana-5267	123	10	uba	uba	PROPN
cana-5267	123	11	)	)	PUNCT
cana-5267	123	12	related	relate	VERB
cana-5267	123	13	not	not	PART
cana-5267	123	14	only	only	ADV
cana-5267	123	15	to	to	PART
cana-5267	123	16	prompt	prompt	VERB
cana-5267	123	17	injections	injection	NOUN
cana-5267	123	18	but	but	CCONJ
cana-5267	123	19	also	also	ADV
cana-5267	123	20	data	data	VERB
cana-5267	123	21	exfiltration	exfiltration	NOUN
cana-5267	123	22	masquerading	masquerade	VERB
cana-5267	123	23	as	as	ADP
cana-5267	123	24	benign	benign	ADJ
cana-5267	123	25	interactions	interaction	NOUN
cana-5267	123	26	.	.	PUNCT
cana-5267	124	1	this	this	DET
cana-5267	124	2	layer	layer	NOUN
cana-5267	124	3	’s	’s	PART
cana-5267	124	4	preprocessing	preprocessing	NOUN
cana-5267	124	5	includes	include	VERB
cana-5267	124	6	data	datum	NOUN
cana-5267	124	7	cleaning	cleaning	NOUN
cana-5267	124	8	and	and	CCONJ
cana-5267	124	9	feature	feature	NOUN
cana-5267	124	10	extraction	extraction	NOUN
cana-5267	124	11	for	for	ADP
cana-5267	124	12	structured	structured	ADJ
cana-5267	124	13	security	security	NOUN
cana-5267	124	14	data	datum	NOUN
cana-5267	124	15	.	.	PUNCT
cana-5267	125	1	the	the	DET
cana-5267	125	2	data	datum	NOUN
cana-5267	125	3	arriving	arrive	VERB
cana-5267	125	4	from	from	ADP
cana-5267	125	5	different	different	ADJ
cana-5267	125	6	verticals	vertical	NOUN
cana-5267	125	7	is	be	AUX
cana-5267	125	8	automatically	automatically	ADV
cana-5267	125	9	parsed	parse	VERB
cana-5267	125	10	,	,	PUNCT
cana-5267	125	11	classified	classified	ADJ
cana-5267	125	12	,	,	PUNCT
cana-5267	125	13	and	and	CCONJ
cana-5267	125	14	categorized	categorize	VERB
cana-5267	125	15	in	in	ADP
cana-5267	125	16	real	real	ADJ
cana-5267	125	17	time	time	NOUN
cana-5267	125	18	.	.	PUNCT
cana-5267	126	1	in	in	ADP
cana-5267	126	2	addition	addition	NOUN
cana-5267	126	3	,	,	PUNCT
cana-5267	126	4	potential	potential	ADJ
cana-5267	126	5	security	security	NOUN
cana-5267	126	6	threats	threat	NOUN
cana-5267	126	7	are	be	AUX
cana-5267	126	8	separated	separate	VERB
cana-5267	126	9	from	from	ADP
cana-5267	126	10	genuine	genuine	ADJ
cana-5267	126	11	user	user	NOUN
cana-5267	126	12	actions	action	NOUN
cana-5267	126	13	.	.	PUNCT
cana-5267	127	1	in	in	ADP
cana-5267	127	2	aid	aid	NOUN
cana-5267	127	3	of	of	ADP
cana-5267	127	4	llm	llm	PROPN
cana-5267	127	5	threat	threat	NOUN
cana-5267	127	6	detection	detection	NOUN
cana-5267	127	7	and	and	CCONJ
cana-5267	127	8	risk	risk	NOUN
cana-5267	127	9	assessment	assessment	NOUN
cana-5267	127	10	models	model	NOUN
cana-5267	127	11	,	,	PUNCT
cana-5267	127	12	this	this	DET
cana-5267	127	13	layer	layer	NOUN
cana-5267	127	14	is	be	AUX
cana-5267	127	15	capable	capable	ADJ
cana-5267	127	16	of	of	ADP
cana-5267	127	17	accuracy	accuracy	NOUN
cana-5267	127	18	engineering	engineering	NOUN
cana-5267	127	19	data	datum	NOUN
cana-5267	127	20	to	to	PART
cana-5267	127	21	preempt	preempt	VERB
cana-5267	127	22	cyber	cyber	NOUN
cana-5267	127	23	threats	threat	NOUN
cana-5267	127	24	at	at	ADP
cana-5267	127	25	ai	ai	ADV
cana-5267	127	26	-	-	PUNCT
cana-5267	127	27	driven	drive	VERB
cana-5267	127	28	enterprise	enterprise	NOUN
cana-5267	127	29	application	application	NOUN
cana-5267	127	30	security	security	NOUN
cana-5267	127	31	.	.	PUNCT
cana-5267	128	1	threat	threat	NOUN
cana-5267	128	2	detection	detection	NOUN
cana-5267	128	3	module	module	NOUN
cana-5267	128	4	:	:	PUNCT
cana-5267	128	5	for	for	ADP
cana-5267	128	6	llm	llm	NOUN
cana-5267	128	7	system	system	NOUN
cana-5267	128	8	integration	integration	NOUN
cana-5267	128	9	automated	automate	VERB
cana-5267	128	10	threat	threat	NOUN
cana-5267	128	11	modeling	modeling	NOUN
cana-5267	128	12	,	,	PUNCT
cana-5267	128	13	the	the	DET
cana-5267	128	14	threat	threat	NOUN
cana-5267	128	15	detection	detection	NOUN
cana-5267	128	16	module	module	NOUN
cana-5267	128	17	is	be	AUX
cana-5267	128	18	the	the	DET
cana-5267	128	19	most	most	ADV
cana-5267	128	20	vital	vital	ADJ
cana-5267	128	21	part	part	NOUN
cana-5267	128	22	because	because	SCONJ
cana-5267	128	23	it	it	PRON
cana-5267	128	24	identifies	identify	VERB
cana-5267	128	25	the	the	DET
cana-5267	128	26	security	security	NOUN
cana-5267	128	27	issues	issue	NOUN
cana-5267	128	28	,	,	PUNCT
cana-5267	128	29	adversary	adversary	NOUN
cana-5267	128	30	-	-	PUNCT
cana-5267	128	31	ship	ship	NOUN
cana-5267	128	32	,	,	PUNCT
cana-5267	128	33	and	and	CCONJ
cana-5267	128	34	anomalies	anomaly	NOUN
cana-5267	128	35	in	in	ADP
cana-5267	128	36	llm	llm	NOUN
cana-5267	128	37	-	-	PUNCT
cana-5267	128	38	based	base	VERB
cana-5267	128	39	applications	application	NOUN
cana-5267	128	40	.	.	PUNCT
cana-5267	129	1	it	it	PRON
cana-5267	129	2	applies	apply	VERB
cana-5267	129	3	ml	ml	ADP
cana-5267	129	4	algorithms	algorithm	NOUN
cana-5267	129	5	,	,	PUNCT
cana-5267	129	6	rule	rule	NOUN
cana-5267	129	7	-	-	PUNCT
cana-5267	129	8	based	base	VERB
cana-5267	129	9	threat	threat	NOUN
cana-5267	129	10	analysis	analysis	NOUN
cana-5267	129	11	,	,	PUNCT
cana-5267	129	12	behavioral	behavioral	ADJ
cana-5267	129	13	analytics	analytic	NOUN
cana-5267	129	14	,	,	PUNCT
cana-5267	129	15	and	and	CCONJ
cana-5267	129	16	other	other	ADJ
cana-5267	129	17	means	mean	NOUN
cana-5267	129	18	to	to	ADP
cana-5267	129	19	real	real	ADJ
cana-5267	129	20	-	-	PUNCT
cana-5267	129	21	time	time	NOUN
cana-5267	129	22	threat	threat	NOUN
cana-5267	129	23	detection	detection	NOUN
cana-5267	129	24	.	.	PUNCT
cana-5267	130	1	stride	stride	NOUN
cana-5267	130	2	(	(	PUNCT
cana-5267	130	3	spoofing	spoofing	NOUN
cana-5267	130	4	,	,	PUNCT
cana-5267	130	5	tampering	tampering	NOUN
cana-5267	130	6	,	,	PUNCT
cana-5267	130	7	repudiation	repudiation	NOUN
cana-5267	130	8	,	,	PUNCT
cana-5267	130	9	information	information	NOUN
cana-5267	130	10	disclosure	disclosure	NOUN
cana-5267	130	11	,	,	PUNCT
cana-5267	130	12	denial	denial	NOUN
cana-5267	130	13	of	of	ADP
cana-5267	130	14	service	service	NOUN
cana-5267	130	15	,	,	PUNCT
cana-5267	130	16	elevation	elevation	NOUN
cana-5267	130	17	of	of	ADP
cana-5267	130	18	privilege	privilege	NOUN
cana-5267	130	19	)	)	PUNCT
cana-5267	130	20	and	and	CCONJ
cana-5267	130	21	dread	dread	NOUN
cana-5267	130	22	(	(	PUNCT
cana-5267	130	23	damage	damage	NOUN
cana-5267	130	24	potential	potential	NOUN
cana-5267	130	25	,	,	PUNCT
cana-5267	130	26	reproducibility	reproducibility	NOUN
cana-5267	130	27	,	,	PUNCT
cana-5267	130	28	exploitability	exploitability	NOUN
cana-5267	130	29	,	,	PUNCT
cana-5267	130	30	affected	affect	VERB
cana-5267	130	31	users	user	NOUN
cana-5267	130	32	,	,	PUNCT
cana-5267	130	33	discoverability	discoverability	NOUN
cana-5267	130	34	)	)	PUNCT
cana-5267	130	35	frameworks	framework	NOUN
cana-5267	130	36	are	be	AUX
cana-5267	130	37	integrated	integrate	VERB
cana-5267	130	38	for	for	ADP
cana-5267	130	39	risk	risk	NOUN
cana-5267	130	40	systematic	systematic	ADJ
cana-5267	130	41	securities	security	NOUN
cana-5267	130	42	goals	goal	NOUN
cana-5267	130	43	assessment	assessment	NOUN
cana-5267	130	44	.	.	PUNCT
cana-5267	131	1	the	the	DET
cana-5267	131	2	system	system	NOUN
cana-5267	131	3	assigns	assign	VERB
cana-5267	131	4	risk	risk	NOUN
cana-5267	131	5	scores	score	NOUN
cana-5267	131	6	and	and	CCONJ
cana-5267	131	7	prioritizes	prioritize	VERB
cana-5267	131	8	security	security	NOUN
cana-5267	131	9	issues	issue	NOUN
cana-5267	131	10	based	base	VERB
cana-5267	131	11	on	on	ADP
cana-5267	131	12	these	these	DET
cana-5267	131	13	models	model	NOUN
cana-5267	131	14	’	'	PUNCT
cana-5267	131	15	detected	detect	VERB
cana-5267	131	16	threats	threat	NOUN
cana-5267	131	17	.	.	PUNCT
cana-5267	132	1	moreover	moreover	ADV
cana-5267	132	2	,	,	PUNCT
cana-5267	132	3	hostile	hostile	ADJ
cana-5267	132	4	prompt	prompt	ADJ
cana-5267	132	5	injection	injection	NOUN
cana-5267	132	6	,	,	PUNCT
cana-5267	132	7	unauthorized	unauthorized	ADJ
cana-5267	132	8	api	api	NOUN
cana-5267	132	9	calls	call	NOUN
cana-5267	132	10	,	,	PUNCT
cana-5267	132	11	and	and	CCONJ
cana-5267	132	12	adversarial	adversarial	ADJ
cana-5267	132	13	attacks	attack	NOUN
cana-5267	132	14	on	on	ADP
cana-5267	132	15	llm	llm	PROPN
cana-5267	132	16	systems	system	NOUN
cana-5267	132	17	are	be	AUX
cana-5267	132	18	exposed	expose	VERB
cana-5267	132	19	through	through	ADP
cana-5267	132	20	nlpbased	nlpbase	VERB
cana-5267	132	21	anomaly	anomaly	PROPN
cana-5267	132	22	detection	detection	NOUN
cana-5267	132	23	.	.	PUNCT
cana-5267	133	1	supervised	supervised	ADJ
cana-5267	133	2	and	and	CCONJ
cana-5267	133	3	unsupervised	unsupervised	ADJ
cana-5267	133	4	increases	increase	NOUN
cana-5267	133	5	threat	threat	NOUN
cana-5267	133	6	detection	detection	NOUN
cana-5267	133	7	efficacy	efficacy	NOUN
cana-5267	133	8	through	through	ADP
cana-5267	133	9	hybrid	hybrid	ADJ
cana-5267	133	10	models	model	NOUN
cana-5267	133	11	.	.	PUNCT
cana-5267	134	1	security	security	NOUN
cana-5267	134	2	datasets	dataset	NOUN
cana-5267	134	3	contain	contain	VERB
cana-5267	134	4	categorized	categorize	VERB
cana-5267	134	5	data	datum	NOUN
cana-5267	134	6	,	,	PUNCT
cana-5267	134	7	but	but	CCONJ
cana-5267	134	8	novel	novel	ADJ
cana-5267	134	9	threats	threat	NOUN
cana-5267	134	10	are	be	AUX
cana-5267	134	11	undetectably	undetectably	ADV
cana-5267	134	12	concealed	conceal	VERB
cana-5267	134	13	in	in	ADP
cana-5267	134	14	normal	normal	ADJ
cana-5267	134	15	behavior	behavior	NOUN
cana-5267	134	16	,	,	PUNCT
cana-5267	134	17	so	so	ADV
cana-5267	134	18	anomaly	anomaly	VERB
cana-5267	134	19	analysis	analysis	NOUN
cana-5267	134	20	uncovers	uncover	NOUN
cana-5267	134	21	them	they	PRON
cana-5267	134	22	through	through	ADP
cana-5267	134	23	unsupervised	unsupervised	ADJ
cana-5267	134	24	learning	learning	NOUN
cana-5267	134	25	.	.	PUNCT
cana-5267	135	1	automated	automate	VERB
cana-5267	135	2	threat	threat	NOUN
cana-5267	135	3	mitigation	mitigation	NOUN
cana-5267	135	4	is	be	AUX
cana-5267	135	5	achieved	achieve	VERB
cana-5267	135	6	by	by	ADP
cana-5267	135	7	continuous	continuous	ADJ
cana-5267	135	8	monitoring	monitoring	NOUN
cana-5267	135	9	of	of	ADP
cana-5267	135	10	user	user	NOUN
cana-5267	135	11	interactions	interaction	NOUN
cana-5267	135	12	,	,	PUNCT
cana-5267	135	13	api	api	NOUN
cana-5267	135	14	calls	call	NOUN
cana-5267	135	15	,	,	PUNCT
cana-5267	135	16	and	and	CCONJ
cana-5267	135	17	model	model	NOUN
cana-5267	135	18	responses	response	NOUN
cana-5267	135	19	by	by	ADP
cana-5267	135	20	the	the	DET
cana-5267	135	21	threat	threat	NOUN
cana-5267	135	22	detection	detection	NOUN
cana-5267	135	23	module	module	NOUN
cana-5267	135	24	.	.	PUNCT
cana-5267	136	1	analytics	analytic	NOUN
cana-5267	136	2	driven	drive	VERB
cana-5267	136	3	by	by	ADP
cana-5267	136	4	ai	ai	PRON
cana-5267	136	5	increase	increase	NOUN
cana-5267	136	6	security	security	NOUN
cana-5267	136	7	posture	posture	NOUN
cana-5267	136	8	,	,	PUNCT
cana-5267	136	9	communications	communication	NOUN
cana-5267	136	10	on	on	ADP
cana-5267	136	11	applied	apply	VERB
cana-5267	136	12	nonlinear	nonlinear	ADJ
cana-5267	136	13	analysis	analysis	NOUN
cana-5267	136	14	issn	issn	NOUN
cana-5267	136	15	:	:	PUNCT
cana-5267	136	16	1074	1074	NUM
cana-5267	136	17	-	-	PUNCT
cana-5267	136	18	133x	133x	NUM
cana-5267	136	19	vol	vol	VERB
cana-5267	136	20	32	32	NUM
cana-5267	136	21	no	no	NOUN
cana-5267	136	22	.	.	PUNCT
cana-5267	137	1	10s	10	NOUN
cana-5267	137	2	(	(	PUNCT
cana-5267	137	3	2025	2025	NUM
cana-5267	137	4	)	)	PUNCT
cana-5267	137	5	1664	1664	NUM
cana-5267	137	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5267	137	7	mitigate	mitigate	VERB
cana-5267	137	8	false	false	ADJ
cana-5267	137	9	alerts	alert	NOUN
cana-5267	137	10	,	,	PUNCT
cana-5267	137	11	and	and	CCONJ
cana-5267	137	12	improve	improve	VERB
cana-5267	137	13	risk	risk	NOUN
cana-5267	137	14	automation	automation	NOUN
cana-5267	137	15	assessment	assessment	NOUN
cana-5267	137	16	in	in	ADP
cana-5267	137	17	llm	llm	ADV
cana-5267	137	18	-	-	PUNCT
cana-5267	137	19	integrated	integrate	VERB
cana-5267	137	20	applications	application	NOUN
cana-5267	137	21	.	.	PUNCT
cana-5267	138	1	and	and	CCONJ
cana-5267	138	2	these	these	DET
cana-5267	138	3	features	feature	NOUN
cana-5267	138	4	bolster	bolster	VERB
cana-5267	138	5	the	the	DET
cana-5267	138	6	system	system	NOUN
cana-5267	138	7	against	against	ADP
cana-5267	138	8	advanced	advanced	ADJ
cana-5267	138	9	persistent	persistent	ADJ
cana-5267	138	10	threats	threat	NOUN
cana-5267	138	11	in	in	ADP
cana-5267	138	12	cybersecurity	cybersecurity	NOUN
cana-5267	138	13	.	.	PUNCT
cana-5267	139	1	risk	risk	NOUN
cana-5267	139	2	assessment	assessment	NOUN
cana-5267	139	3	engine	engine	NOUN
cana-5267	139	4	:	:	PUNCT
cana-5267	139	5	the	the	DET
cana-5267	139	6	idea	idea	NOUN
cana-5267	139	7	behind	behind	ADP
cana-5267	139	8	this	this	DET
cana-5267	139	9	research	research	NOUN
cana-5267	139	10	is	be	AUX
cana-5267	139	11	to	to	PART
cana-5267	139	12	integrate	integrate	VERB
cana-5267	139	13	a	a	DET
cana-5267	139	14	threat	threat	NOUN
cana-5267	139	15	modeling	model	VERB
cana-5267	139	16	framework	framework	NOUN
cana-5267	139	17	with	with	ADP
cana-5267	139	18	risk	risk	NOUN
cana-5267	139	19	assessment	assessment	NOUN
cana-5267	139	20	to	to	PART
cana-5267	139	21	identify	identify	VERB
cana-5267	139	22	llm	llm	NOUN
cana-5267	139	23	models	model	NOUN
cana-5267	139	24	and	and	CCONJ
cana-5267	139	25	their	their	PRON
cana-5267	139	26	systems	system	NOUN
cana-5267	139	27	’	'	PUNCT
cana-5267	139	28	integration	integration	NOUN
cana-5267	139	29	risks	risk	NOUN
cana-5267	139	30	.	.	PUNCT
cana-5267	140	1	the	the	DET
cana-5267	140	2	risk	risk	NOUN
cana-5267	140	3	scoring	scoring	NOUN
cana-5267	140	4	and	and	CCONJ
cana-5267	140	5	prioritization	prioritization	NOUN
cana-5267	140	6	process	process	NOUN
cana-5267	140	7	is	be	AUX
cana-5267	140	8	likely	likely	ADJ
cana-5267	140	9	a	a	DET
cana-5267	140	10	part	part	NOUN
cana-5267	140	11	of	of	ADP
cana-5267	140	12	the	the	DET
cana-5267	140	13	risk	risk	NOUN
cana-5267	140	14	assessment	assessment	NOUN
cana-5267	140	15	engine	engine	NOUN
cana-5267	140	16	in	in	ADP
cana-5267	140	17	the	the	DET
cana-5267	140	18	automated	automate	VERB
cana-5267	140	19	threat	threat	NOUN
cana-5267	140	20	modeling	model	VERB
cana-5267	140	21	framework	framework	NOUN
cana-5267	140	22	.	.	PUNCT
cana-5267	141	1	as	as	SCONJ
cana-5267	141	2	mentioned	mention	VERB
cana-5267	141	3	,	,	PUNCT
cana-5267	141	4	fuzzy	fuzzy	ADJ
cana-5267	141	5	logic	logic	NOUN
cana-5267	141	6	,	,	PUNCT
cana-5267	141	7	reinforcement	reinforcement	NOUN
cana-5267	141	8	learning	learning	NOUN
cana-5267	141	9	,	,	PUNCT
cana-5267	141	10	and	and	CCONJ
cana-5267	141	11	behavioral	behavioral	ADJ
cana-5267	141	12	analytics	analytic	NOUN
cana-5267	141	13	facilitate	facilitate	VERB
cana-5267	141	14	the	the	DET
cana-5267	141	15	dynamic	dynamic	ADJ
cana-5267	141	16	security	security	NOUN
cana-5267	141	17	risk	risk	NOUN
cana-5267	141	18	evaluation	evaluation	NOUN
cana-5267	141	19	in	in	ADP
cana-5267	141	20	real	real	ADJ
cana-5267	141	21	time	time	NOUN
cana-5267	141	22	.	.	PUNCT
cana-5267	142	1	making	make	VERB
cana-5267	142	2	use	use	NOUN
cana-5267	142	3	of	of	ADP
cana-5267	142	4	machine	machine	NOUN
cana-5267	142	5	learning	learning	NOUN
cana-5267	142	6	based	base	VERB
cana-5267	142	7	classifiers	classifier	NOUN
cana-5267	142	8	,	,	PUNCT
cana-5267	142	9	this	this	DET
cana-5267	142	10	engine	engine	NOUN
cana-5267	142	11	assesses	assess	VERB
cana-5267	142	12	the	the	DET
cana-5267	142	13	threat	threat	NOUN
cana-5267	142	14	into	into	ADP
cana-5267	142	15	different	different	ADJ
cana-5267	142	16	risk	risk	NOUN
cana-5267	142	17	categories	category	NOUN
cana-5267	142	18	ensuring	ensure	VERB
cana-5267	142	19	a	a	DET
cana-5267	142	20	data	datum	NOUN
cana-5267	142	21	driven	drive	VERB
cana-5267	142	22	adaptive	adaptive	ADJ
cana-5267	142	23	approach	approach	NOUN
cana-5267	142	24	to	to	ADP
cana-5267	142	25	threat	threat	NOUN
cana-5267	142	26	evaluation	evaluation	NOUN
cana-5267	142	27	.	.	PUNCT
cana-5267	143	1	threat	threat	NOUN
cana-5267	143	2	evaluation	evaluation	NOUN
cana-5267	143	3	is	be	AUX
cana-5267	143	4	further	far	ADV
cana-5267	143	5	enhanced	enhance	VERB
cana-5267	143	6	by	by	ADP
cana-5267	143	7	integrating	integrate	VERB
cana-5267	143	8	stride	stride	NOUN
cana-5267	143	9	and	and	CCONJ
cana-5267	143	10	dread	dread	NOUN
cana-5267	143	11	frameworks	framework	NOUN
cana-5267	143	12	enabling	enable	VERB
cana-5267	143	13	the	the	DET
cana-5267	143	14	engine	engine	NOUN
cana-5267	143	15	to	to	PART
cana-5267	143	16	give	give	VERB
cana-5267	143	17	a	a	DET
cana-5267	143	18	quantitative	quantitative	ADJ
cana-5267	143	19	assessment	assessment	NOUN
cana-5267	143	20	of	of	ADP
cana-5267	143	21	the	the	DET
cana-5267	143	22	evaluated	evaluated	ADJ
cana-5267	143	23	security	security	NOUN
cana-5267	143	24	threats	threat	NOUN
cana-5267	143	25	.	.	PUNCT
cana-5267	144	1	apart	apart	ADV
cana-5267	144	2	from	from	ADP
cana-5267	144	3	these	these	PRON
cana-5267	144	4	,	,	PUNCT
cana-5267	144	5	adversary	adversary	NOUN
cana-5267	144	6	supplied	supply	VERB
cana-5267	144	7	inputs	input	NOUN
cana-5267	144	8	,	,	PUNCT
cana-5267	144	9	injections	injection	NOUN
cana-5267	144	10	,	,	PUNCT
cana-5267	144	11	and	and	CCONJ
cana-5267	144	12	illegal	illegal	ADJ
cana-5267	144	13	attempts	attempt	NOUN
cana-5267	144	14	to	to	PART
cana-5267	144	15	access	access	VERB
cana-5267	144	16	the	the	DET
cana-5267	144	17	apis	apis	NOUN
cana-5267	144	18	provided	provide	VERB
cana-5267	144	19	by	by	ADP
cana-5267	144	20	the	the	DET
cana-5267	144	21	llm	llm	PROPN
cana-5267	144	22	’s	’s	PART
cana-5267	144	23	are	be	AUX
cana-5267	144	24	monitored	monitor	VERB
cana-5267	144	25	and	and	CCONJ
cana-5267	144	26	analyzed	analyze	VERB
cana-5267	144	27	as	as	ADP
cana-5267	144	28	users	user	NOUN
cana-5267	144	29	through	through	ADP
cana-5267	144	30	advanced	advanced	ADJ
cana-5267	144	31	anomaly	anomaly	NOUN
cana-5267	144	32	detection	detection	NOUN
cana-5267	144	33	systems	system	NOUN
cana-5267	144	34	to	to	PART
cana-5267	144	35	capture	capture	VERB
cana-5267	144	36	unauthorized	unauthorized	ADJ
cana-5267	144	37	actions	action	NOUN
cana-5267	144	38	.	.	PUNCT
cana-5267	145	1	the	the	DET
cana-5267	145	2	results	result	NOUN
cana-5267	145	3	from	from	ADP
cana-5267	145	4	the	the	DET
cana-5267	145	5	risk	risk	NOUN
cana-5267	145	6	assessment	assessment	NOUN
cana-5267	145	7	are	be	AUX
cana-5267	145	8	utilized	utilize	VERB
cana-5267	145	9	by	by	ADP
cana-5267	145	10	the	the	DET
cana-5267	145	11	automated	automate	VERB
cana-5267	145	12	mitigation	mitigation	NOUN
cana-5267	145	13	system	system	NOUN
cana-5267	145	14	which	which	PRON
cana-5267	145	15	exercises	exercise	VERB
cana-5267	145	16	control	control	VERB
cana-5267	145	17	over	over	ADP
cana-5267	145	18	the	the	DET
cana-5267	145	19	system	system	NOUN
cana-5267	145	20	through	through	ADP
cana-5267	145	21	policy	policy	NOUN
cana-5267	145	22	application	application	NOUN
cana-5267	145	23	,	,	PUNCT
cana-5267	145	24	api	api	NOUN
cana-5267	145	25	moderation	moderation	NOUN
cana-5267	145	26	,	,	PUNCT
cana-5267	145	27	and	and	CCONJ
cana-5267	145	28	input	input	NOUN
cana-5267	145	29	filtering	filtering	NOUN
cana-5267	145	30	.	.	PUNCT
cana-5267	146	1	the	the	DET
cana-5267	146	2	continuously	continuously	ADV
cana-5267	146	3	evolving	evolve	VERB
cana-5267	146	4	threats	threat	NOUN
cana-5267	146	5	enhance	enhance	VERB
cana-5267	146	6	the	the	DET
cana-5267	146	7	risk	risk	NOUN
cana-5267	146	8	assessment	assessment	NOUN
cana-5267	146	9	engine	engine	NOUN
cana-5267	146	10	,	,	PUNCT
cana-5267	146	11	improving	improve	VERB
cana-5267	146	12	the	the	DET
cana-5267	146	13	real	real	ADJ
cana-5267	146	14	security	security	NOUN
cana-5267	146	15	monitoring	monitoring	NOUN
cana-5267	146	16	while	while	SCONJ
cana-5267	146	17	enabling	enable	VERB
cana-5267	146	18	llm	llm	PROPN
cana-5267	146	19	applications	application	NOUN
cana-5267	146	20	to	to	PART
cana-5267	146	21	actively	actively	ADV
cana-5267	146	22	respond	respond	VERB
cana-5267	146	23	to	to	ADP
cana-5267	146	24	changing	change	VERB
cana-5267	146	25	cyber	cyber	NOUN
cana-5267	146	26	threats	threat	NOUN
cana-5267	146	27	due	due	ADJ
cana-5267	146	28	to	to	PART
cana-5267	146	29	antinist	antinist	VERB
cana-5267	146	30	and	and	CCONJ
cana-5267	146	31	anti	anti	ADJ
cana-5267	146	32	-	-	ADJ
cana-5267	146	33	gdpr	gdpr	ADJ
cana-5267	146	34	regulation	regulation	NOUN
cana-5267	146	35	compliance	compliance	NOUN
cana-5267	146	36	with	with	ADP
cana-5267	146	37	pre	pre	ADJ
cana-5267	146	38	-	-	ADJ
cana-5267	146	39	established	established	ADJ
cana-5267	146	40	red	red	ADJ
cana-5267	146	41	tape	tape	NOUN
cana-5267	146	42	regulations	regulation	NOUN
cana-5267	146	43	.	.	PUNCT
cana-5267	147	1	automated	automate	VERB
cana-5267	147	2	mitigation	mitigation	NOUN
cana-5267	147	3	system	system	NOUN
cana-5267	147	4	:	:	PUNCT
cana-5267	147	5	the	the	DET
cana-5267	147	6	ams	am	NOUN
cana-5267	147	7	,	,	PUNCT
cana-5267	147	8	or	or	CCONJ
cana-5267	147	9	automated	automate	VERB
cana-5267	147	10	mitigation	mitigation	NOUN
cana-5267	147	11	system	system	NOUN
cana-5267	147	12	,	,	PUNCT
cana-5267	147	13	is	be	AUX
cana-5267	147	14	part	part	NOUN
cana-5267	147	15	of	of	ADP
cana-5267	147	16	the	the	DET
cana-5267	147	17	larger	large	ADJ
cana-5267	147	18	automated	automate	VERB
cana-5267	147	19	threat	threat	NOUN
cana-5267	147	20	modeling	modeling	NOUN
cana-5267	147	21	for	for	ADP
cana-5267	147	22	llm	llm	PROPN
cana-5267	147	23	system	system	NOUN
cana-5267	147	24	integration	integration	NOUN
cana-5267	147	25	.	.	PUNCT
cana-5267	148	1	it	it	PRON
cana-5267	148	2	addresses	address	VERB
cana-5267	148	3	detected	detect	VERB
cana-5267	148	4	threats	threat	NOUN
cana-5267	148	5	in	in	ADP
cana-5267	148	6	real	real	ADJ
cana-5267	148	7	-	-	PUNCT
cana-5267	148	8	time	time	NOUN
cana-5267	148	9	via	via	ADP
cana-5267	148	10	adaptive	adaptive	ADJ
cana-5267	148	11	security	security	NOUN
cana-5267	148	12	measures	measure	NOUN
cana-5267	148	13	.	.	PUNCT
cana-5267	149	1	after	after	ADP
cana-5267	149	2	the	the	DET
cana-5267	149	3	risk	risk	NOUN
cana-5267	149	4	assessment	assessment	NOUN
cana-5267	149	5	engine	engine	NOUN
cana-5267	149	6	detects	detect	NOUN
cana-5267	149	7	and	and	CCONJ
cana-5267	149	8	evaluates	evaluate	VERB
cana-5267	149	9	a	a	DET
cana-5267	149	10	threat	threat	NOUN
cana-5267	149	11	,	,	PUNCT
cana-5267	149	12	the	the	DET
cana-5267	149	13	ams	ams	NOUN
cana-5267	149	14	is	be	AUX
cana-5267	149	15	capable	capable	ADJ
cana-5267	149	16	of	of	ADP
cana-5267	149	17	dynamically	dynamically	ADV
cana-5267	149	18	creating	create	VERB
cana-5267	149	19	,	,	PUNCT
cana-5267	149	20	enforcing	enforce	VERB
cana-5267	149	21	,	,	PUNCT
cana-5267	149	22	and	and	CCONJ
cana-5267	149	23	adjusting	adjust	VERB
cana-5267	149	24	policies	policy	NOUN
cana-5267	149	25	to	to	PART
cana-5267	149	26	secure	secure	VERB
cana-5267	149	27	any	any	DET
cana-5267	149	28	windows	window	NOUN
cana-5267	149	29	of	of	ADP
cana-5267	149	30	exploitation	exploitation	NOUN
cana-5267	149	31	.	.	PUNCT
cana-5267	150	1	it	it	PRON
cana-5267	150	2	utilizes	utilize	VERB
cana-5267	150	3	security	security	NOUN
cana-5267	150	4	automation	automation	NOUN
cana-5267	150	5	through	through	ADP
cana-5267	150	6	machine	machine	NOUN
cana-5267	150	7	learning	learn	VERB
cana-5267	150	8	to	to	PART
cana-5267	150	9	combat	combat	VERB
cana-5267	150	10	prompt	prompt	ADJ
cana-5267	150	11	injection	injection	NOUN
cana-5267	150	12	attacks	attack	NOUN
cana-5267	150	13	,	,	PUNCT
cana-5267	150	14	data	data	NOUN
cana-5267	150	15	poisoning	poisoning	NOUN
cana-5267	150	16	,	,	PUNCT
cana-5267	150	17	adversarial	adversarial	ADJ
cana-5267	150	18	manipulation	manipulation	NOUN
cana-5267	150	19	,	,	PUNCT
cana-5267	150	20	and	and	CCONJ
cana-5267	150	21	attempts	attempt	VERB
cana-5267	150	22	at	at	ADP
cana-5267	150	23	unauthorized	unauthorized	ADJ
cana-5267	150	24	access	access	NOUN
cana-5267	150	25	.	.	PUNCT
cana-5267	151	1	by	by	ADP
cana-5267	151	2	employing	employ	VERB
cana-5267	151	3	policy	policy	NOUN
cana-5267	151	4	-	-	PUNCT
cana-5267	151	5	based	base	VERB
cana-5267	151	6	enforcement	enforcement	NOUN
cana-5267	151	7	,	,	PUNCT
cana-5267	151	8	the	the	DET
cana-5267	151	9	ams	am	NOUN
cana-5267	151	10	can	can	AUX
cana-5267	151	11	change	change	VERB
cana-5267	151	12	api	api	NOUN
cana-5267	151	13	call	call	NOUN
cana-5267	151	14	quotas	quota	NOUN
cana-5267	151	15	,	,	PUNCT
cana-5267	151	16	modify	modify	VERB
cana-5267	151	17	access	access	NOUN
cana-5267	151	18	restrictions	restriction	NOUN
cana-5267	151	19	,	,	PUNCT
cana-5267	151	20	apply	apply	VERB
cana-5267	151	21	filtering	filtering	NOUN
cana-5267	151	22	of	of	ADP
cana-5267	151	23	adversarial	adversarial	ADJ
cana-5267	151	24	inputs	input	NOUN
cana-5267	151	25	,	,	PUNCT
cana-5267	151	26	and	and	CCONJ
cana-5267	151	27	sever	sever	VERB
cana-5267	151	28	high	high	ADJ
cana-5267	151	29	-	-	PUNCT
cana-5267	151	30	risk	risk	NOUN
cana-5267	151	31	interactions	interaction	NOUN
cana-5267	151	32	to	to	ADP
cana-5267	151	33	lower	low	ADJ
cana-5267	151	34	threats	threat	NOUN
cana-5267	151	35	.	.	PUNCT
cana-5267	152	1	moreover	moreover	ADV
cana-5267	152	2	,	,	PUNCT
cana-5267	152	3	it	it	PRON
cana-5267	152	4	integrates	integrate	VERB
cana-5267	152	5	behavioral	behavioral	ADJ
cana-5267	152	6	analysis	analysis	NOUN
cana-5267	152	7	aimed	aim	VERB
cana-5267	152	8	at	at	ADP
cana-5267	152	9	capturing	capture	VERB
cana-5267	152	10	suspicious	suspicious	ADJ
cana-5267	152	11	user	user	NOUN
cana-5267	152	12	interaction	interaction	NOUN
cana-5267	152	13	patterns	pattern	NOUN
cana-5267	152	14	and	and	CCONJ
cana-5267	152	15	altering	alter	VERB
cana-5267	152	16	corresponding	corresponding	ADJ
cana-5267	152	17	strategies	strategy	NOUN
cana-5267	152	18	.	.	PUNCT
cana-5267	153	1	ams	ams	PROPN
cana-5267	153	2	’s	’s	PART
cana-5267	153	3	self	self	NOUN
cana-5267	153	4	-	-	PUNCT
cana-5267	153	5	learning	learning	NOUN
cana-5267	153	6	,	,	PUNCT
cana-5267	153	7	or	or	CCONJ
cana-5267	153	8	reinforcement	reinforcement	NOUN
cana-5267	153	9	learning	learning	NOUN
cana-5267	153	10	algorithms	algorithm	NOUN
cana-5267	153	11	,	,	PUNCT
cana-5267	153	12	equips	equip	VERB
cana-5267	153	13	it	it	PRON
cana-5267	153	14	with	with	ADP
cana-5267	153	15	the	the	DET
cana-5267	153	16	ability	ability	NOUN
cana-5267	153	17	to	to	PART
cana-5267	153	18	autonomously	autonomously	ADV
cana-5267	153	19	and	and	CCONJ
cana-5267	153	20	continuously	continuously	ADV
cana-5267	153	21	optimize	optimize	VERB
cana-5267	153	22	mitigation	mitigation	NOUN
cana-5267	153	23	frameworks	framework	NOUN
cana-5267	153	24	in	in	ADP
cana-5267	153	25	response	response	NOUN
cana-5267	153	26	to	to	ADP
cana-5267	153	27	shifting	shift	VERB
cana-5267	153	28	attack	attack	NOUN
cana-5267	153	29	rhythms	rhythm	NOUN
cana-5267	153	30	,	,	PUNCT
cana-5267	153	31	enabling	enable	VERB
cana-5267	153	32	effective	effective	ADJ
cana-5267	153	33	attack	attack	NOUN
cana-5267	153	34	adaptation	adaptation	NOUN
cana-5267	153	35	.	.	PUNCT
cana-5267	154	1	ams	am	NOUN
cana-5267	154	2	integrates	integrate	NOUN
cana-5267	154	3	with	with	ADP
cana-5267	154	4	siem	siem	NOUN
cana-5267	154	5	,	,	PUNCT
cana-5267	154	6	or	or	CCONJ
cana-5267	154	7	security	security	NOUN
cana-5267	154	8	information	information	NOUN
cana-5267	154	9	and	and	CCONJ
cana-5267	154	10	event	event	NOUN
cana-5267	154	11	management	management	NOUN
cana-5267	154	12	,	,	PUNCT
cana-5267	154	13	fostering	foster	VERB
cana-5267	154	14	cross	cross	ADJ
cana-5267	154	15	-	-	ADJ
cana-5267	154	16	domain	domain	ADJ
cana-5267	154	17	enterprise	enterprise	NOUN
cana-5267	154	18	security	security	NOUN
cana-5267	154	19	monitoring	monitoring	NOUN
cana-5267	154	20	.	.	PUNCT
cana-5267	155	1	these	these	DET
cana-5267	155	2	features	feature	NOUN
cana-5267	155	3	aid	aid	VERB
cana-5267	155	4	the	the	DET
cana-5267	155	5	augmentation	augmentation	NOUN
cana-5267	155	6	of	of	ADP
cana-5267	155	7	manual	manual	ADJ
cana-5267	155	8	interventions	intervention	NOUN
cana-5267	155	9	in	in	ADP
cana-5267	155	10	security	security	NOUN
cana-5267	155	11	breaches	breach	NOUN
cana-5267	155	12	as	as	ADV
cana-5267	155	13	well	well	ADV
cana-5267	155	14	as	as	ADP
cana-5267	155	15	the	the	DET
cana-5267	155	16	automation	automation	NOUN
cana-5267	155	17	of	of	ADP
cana-5267	155	18	response	response	NOUN
cana-5267	155	19	processes	process	NOUN
cana-5267	155	20	.	.	PUNCT
cana-5267	156	1	therefore	therefore	ADV
cana-5267	156	2	,	,	PUNCT
cana-5267	156	3	ams	ams	PROPN
cana-5267	156	4	’s	’s	PART
cana-5267	156	5	mitigation	mitigation	NOUN
cana-5267	156	6	features	feature	NOUN
cana-5267	156	7	enhance	enhance	VERB
cana-5267	156	8	the	the	DET
cana-5267	156	9	efficiency	efficiency	NOUN
cana-5267	156	10	,	,	PUNCT
cana-5267	156	11	reduce	reduce	VERB
cana-5267	156	12	the	the	DET
cana-5267	156	13	overreliance	overreliance	NOUN
cana-5267	156	14	of	of	ADP
cana-5267	156	15	manual	manual	ADJ
cana-5267	156	16	actions	action	NOUN
cana-5267	156	17	on	on	ADP
cana-5267	156	18	ai	ai	ADJ
cana-5267	156	19	-	-	PUNCT
cana-5267	156	20	integrated	integrate	VERB
cana-5267	156	21	systems	system	NOUN
cana-5267	156	22	,	,	PUNCT
cana-5267	156	23	and	and	CCONJ
cana-5267	156	24	diminish	diminish	VERB
cana-5267	156	25	false	false	ADJ
cana-5267	156	26	alarms	alarm	NOUN
cana-5267	156	27	.	.	PUNCT
cana-5267	157	1	these	these	DET
cana-5267	157	2	features	feature	NOUN
cana-5267	157	3	fortify	fortify	VERB
cana-5267	157	4	the	the	DET
cana-5267	157	5	position	position	NOUN
cana-5267	157	6	of	of	ADP
cana-5267	157	7	ams	am	NOUN
cana-5267	157	8	as	as	ADP
cana-5267	157	9	central	central	ADJ
cana-5267	157	10	to	to	ADP
cana-5267	157	11	the	the	DET
cana-5267	157	12	protection	protection	NOUN
cana-5267	157	13	of	of	ADP
cana-5267	157	14	enterprise	enterprise	NOUN
cana-5267	157	15	systems	system	NOUN
cana-5267	157	16	and	and	CCONJ
cana-5267	157	17	ai	ai	VERB
cana-5267	157	18	financial	financial	ADJ
cana-5267	157	19	systems	system	NOUN
cana-5267	157	20	.	.	PUNCT
cana-5267	158	1	enterprise	enterprise	NOUN
cana-5267	158	2	integration	integration	NOUN
cana-5267	158	3	layer	layer	NOUN
cana-5267	158	4	:	:	PUNCT
cana-5267	158	5	integrating	integrate	VERB
cana-5267	158	6	technologies	technology	NOUN
cana-5267	158	7	as	as	ADP
cana-5267	158	8	a	a	DET
cana-5267	158	9	new	new	ADJ
cana-5267	158	10	layer	layer	NOUN
cana-5267	158	11	(	(	PUNCT
cana-5267	158	12	ill	ill	ADV
cana-5267	158	13	)	)	PUNCT
cana-5267	158	14	allows	allow	VERB
cana-5267	158	15	communication	communication	NOUN
cana-5267	158	16	between	between	ADP
cana-5267	158	17	architectural	architectural	ADJ
cana-5267	158	18	components	component	NOUN
cana-5267	158	19	in	in	ADP
cana-5267	158	20	real	real	ADJ
cana-5267	158	21	time	time	NOUN
cana-5267	158	22	exposing	expose	VERB
cana-5267	158	23	interfaces	interface	NOUN
cana-5267	158	24	to	to	ADP
cana-5267	158	25	existing	exist	VERB
cana-5267	158	26	enterprise	enterprise	NOUN
cana-5267	158	27	security	security	NOUN
cana-5267	158	28	structures	structure	NOUN
cana-5267	158	29	.	.	PUNCT
cana-5267	159	1	the	the	DET
cana-5267	159	2	subsystem	subsystem	NOUN
cana-5267	159	3	enables	enable	VERB
cana-5267	159	4	real	real	ADJ
cana-5267	159	5	time	time	NOUN
cana-5267	159	6	threat	threat	NOUN
cana-5267	159	7	monitoring	monitoring	NOUN
cana-5267	159	8	,	,	PUNCT
cana-5267	159	9	response	response	NOUN
cana-5267	159	10	automation	automation	NOUN
cana-5267	159	11	,	,	PUNCT
cana-5267	159	12	and	and	CCONJ
cana-5267	159	13	integration	integration	NOUN
cana-5267	159	14	with	with	ADP
cana-5267	159	15	siem	siem	PROPN
cana-5267	159	16	,	,	PUNCT
cana-5267	159	17	edr	edr	PROPN
cana-5267	159	18	,	,	PUNCT
cana-5267	159	19	and	and	CCONJ
cana-5267	159	20	casb	casb	NOUN
cana-5267	159	21	security	security	NOUN
cana-5267	159	22	frameworks	framework	NOUN
cana-5267	159	23	.	.	PUNCT
cana-5267	160	1	this	this	DET
cana-5267	160	2	layer	layer	NOUN
cana-5267	160	3	allows	allow	VERB
cana-5267	160	4	security	security	NOUN
cana-5267	160	5	monitoring	monitoring	NOUN
cana-5267	160	6	by	by	ADP
cana-5267	160	7	llms	llm	NOUN
cana-5267	160	8	to	to	PART
cana-5267	160	9	interface	interface	VERB
cana-5267	160	10	with	with	ADP
cana-5267	160	11	enterprise	enterprise	NOUN
cana-5267	160	12	level	level	NOUN
cana-5267	160	13	cybersecurity	cybersecurity	NOUN
cana-5267	160	14	systems	system	NOUN
cana-5267	160	15	making	make	VERB
cana-5267	160	16	communications	communication	NOUN
cana-5267	160	17	on	on	ADP
cana-5267	160	18	applied	apply	VERB
cana-5267	160	19	nonlinear	nonlinear	ADJ
cana-5267	160	20	analysis	analysis	NOUN
cana-5267	160	21	issn	issn	NOUN
cana-5267	160	22	:	:	PUNCT
cana-5267	160	23	1074	1074	NUM
cana-5267	160	24	-	-	PUNCT
cana-5267	160	25	133x	133x	NUM
cana-5267	160	26	vol	vol	VERB
cana-5267	160	27	32	32	NUM
cana-5267	160	28	no	no	NOUN
cana-5267	160	29	.	.	PUNCT
cana-5267	161	1	10s	10	NOUN
cana-5267	161	2	(	(	PUNCT
cana-5267	161	3	2025	2025	NUM
cana-5267	161	4	)	)	PUNCT
cana-5267	161	5	1665	1665	NUM
cana-5267	161	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5267	161	7	centralization	centralization	NOUN
cana-5267	161	8	of	of	ADP
cana-5267	161	9	threat	threat	NOUN
cana-5267	161	10	intelligence	intelligence	NOUN
cana-5267	161	11	,	,	PUNCT
cana-5267	161	12	risk	risk	NOUN
cana-5267	161	13	evaluation	evaluation	NOUN
cana-5267	161	14	,	,	PUNCT
cana-5267	161	15	and	and	CCONJ
cana-5267	161	16	incident	incident	NOUN
cana-5267	161	17	response	response	NOUN
cana-5267	161	18	achievable	achievable	ADJ
cana-5267	161	19	.	.	PUNCT
cana-5267	162	1	through	through	ADP
cana-5267	162	2	apis	apis	PROPN
cana-5267	162	3	,	,	PUNCT
cana-5267	162	4	event	event	NOUN
cana-5267	162	5	driven	drive	VERB
cana-5267	162	6	architectures	architecture	NOUN
cana-5267	162	7	,	,	PUNCT
cana-5267	162	8	and	and	CCONJ
cana-5267	162	9	interoperability	interoperability	NOUN
cana-5267	162	10	standards	standard	NOUN
cana-5267	162	11	incorporated	incorporate	VERB
cana-5267	162	12	in	in	ADP
cana-5267	162	13	this	this	DET
cana-5267	162	14	layer	layer	NOUN
cana-5267	162	15	,	,	PUNCT
cana-5267	162	16	the	the	DET
cana-5267	162	17	enterprise	enterprise	NOUN
cana-5267	162	18	integration	integration	NOUN
cana-5267	162	19	layer	layer	NOUN
cana-5267	162	20	facilitates	facilitate	VERB
cana-5267	162	21	automatic	automatic	ADJ
cana-5267	162	22	triggering	triggering	NOUN
cana-5267	162	23	of	of	ADP
cana-5267	162	24	alerts	alert	NOUN
cana-5267	162	25	,	,	PUNCT
cana-5267	162	26	enforcement	enforcement	NOUN
cana-5267	162	27	of	of	ADP
cana-5267	162	28	security	security	NOUN
cana-5267	162	29	policies	policy	NOUN
cana-5267	162	30	,	,	PUNCT
cana-5267	162	31	and	and	CCONJ
cana-5267	162	32	application	application	NOUN
cana-5267	162	33	of	of	ADP
cana-5267	162	34	active	active	ADJ
cana-5267	162	35	measures	measure	NOUN
cana-5267	162	36	across	across	ADP
cana-5267	162	37	various	various	ADJ
cana-5267	162	38	enterprise	enterprise	NOUN
cana-5267	162	39	level	level	NOUN
cana-5267	162	40	applications	application	NOUN
cana-5267	162	41	.	.	PUNCT
cana-5267	163	1	this	this	DET
cana-5267	163	2	subsystem	subsystem	NOUN
cana-5267	163	3	aids	aid	NOUN
cana-5267	163	4	in	in	ADP
cana-5267	163	5	the	the	DET
cana-5267	163	6	aggregation	aggregation	NOUN
cana-5267	163	7	of	of	ADP
cana-5267	163	8	logs	log	NOUN
cana-5267	163	9	,	,	PUNCT
cana-5267	163	10	behavioral	behavioral	ADJ
cana-5267	163	11	analysis	analysis	NOUN
cana-5267	163	12	,	,	PUNCT
cana-5267	163	13	and	and	CCONJ
cana-5267	163	14	anomaly	anomaly	NOUN
cana-5267	163	15	detection	detection	NOUN
cana-5267	163	16	assuring	assure	VERB
cana-5267	163	17	llm	llm	NOUN
cana-5267	163	18	-	-	PUNCT
cana-5267	163	19	based	base	VERB
cana-5267	163	20	systems	system	NOUN
cana-5267	163	21	are	be	AUX
cana-5267	163	22	protected	protect	VERB
cana-5267	163	23	and	and	CCONJ
cana-5267	163	24	monitored	monitor	VERB
cana-5267	163	25	around	around	ADP
cana-5267	163	26	the	the	DET
cana-5267	163	27	clock	clock	NOUN
cana-5267	163	28	against	against	ADP
cana-5267	163	29	breaches	breach	NOUN
cana-5267	163	30	,	,	PUNCT
cana-5267	163	31	attacks	attack	NOUN
cana-5267	163	32	,	,	PUNCT
cana-5267	163	33	and	and	CCONJ
cana-5267	163	34	access	access	NOUN
cana-5267	163	35	without	without	ADP
cana-5267	163	36	permission	permission	NOUN
cana-5267	163	37	.	.	PUNCT
cana-5267	164	1	correlation	correlation	NOUN
cana-5267	164	2	of	of	ADP
cana-5267	164	3	llm	llm	PROPN
cana-5267	164	4	dominated	dominate	VERB
cana-5267	164	5	security	security	NOUN
cana-5267	164	6	events	event	NOUN
cana-5267	164	7	with	with	ADP
cana-5267	164	8	other	other	ADJ
cana-5267	164	9	cybersecurity	cybersecurity	NOUN
cana-5267	164	10	-	-	PUNCT
cana-5267	164	11	using	use	VERB
cana-5267	164	12	siem	siem	NOUN
cana-5267	164	13	enhances	enhance	VERB
cana-5267	164	14	the	the	DET
cana-5267	164	15	precision	precision	NOUN
cana-5267	164	16	of	of	ADP
cana-5267	164	17	threat	threat	NOUN
cana-5267	164	18	identification	identification	NOUN
cana-5267	164	19	and	and	CCONJ
cana-5267	164	20	improving	improve	VERB
cana-5267	164	21	responses	response	NOUN
cana-5267	164	22	to	to	ADP
cana-5267	164	23	security	security	NOUN
cana-5267	164	24	incidents	incident	NOUN
cana-5267	164	25	.	.	PUNCT
cana-5267	165	1	embedding	embed	VERB
cana-5267	165	2	automated	automate	VERB
cana-5267	165	3	threat	threat	NOUN
cana-5267	165	4	modeling	modeling	NOUN
cana-5267	165	5	powered	power	VERB
cana-5267	165	6	by	by	ADP
cana-5267	165	7	artificial	artificial	ADJ
cana-5267	165	8	intelligence	intelligence	NOUN
cana-5267	165	9	into	into	ADP
cana-5267	165	10	the	the	DET
cana-5267	165	11	enterprise	enterprise	NOUN
cana-5267	165	12	infrastructure	infrastructure	NOUN
cana-5267	165	13	using	use	VERB
cana-5267	165	14	this	this	DET
cana-5267	165	15	layer	layer	NOUN
cana-5267	165	16	optimizes	optimize	VERB
cana-5267	165	17	the	the	DET
cana-5267	165	18	security	security	NOUN
cana-5267	165	19	stance	stance	NOUN
cana-5267	165	20	of	of	ADP
cana-5267	165	21	the	the	DET
cana-5267	165	22	system	system	NOUN
cana-5267	165	23	,	,	PUNCT
cana-5267	165	24	regulatory	regulatory	ADJ
cana-5267	165	25	adherence	adherence	NOUN
cana-5267	165	26	,	,	PUNCT
cana-5267	165	27	and	and	CCONJ
cana-5267	165	28	risk	risk	NOUN
cana-5267	165	29	control	control	NOUN
cana-5267	165	30	strengthening	strengthen	VERB
cana-5267	165	31	the	the	DET
cana-5267	165	32	defenses	defense	NOUN
cana-5267	165	33	against	against	ADP
cana-5267	165	34	evolving	evolve	VERB
cana-5267	165	35	cyber	cyber	NOUN
cana-5267	165	36	attacks	attack	NOUN
cana-5267	165	37	while	while	SCONJ
cana-5267	165	38	making	make	VERB
cana-5267	165	39	llm	llm	PROPN
cana-5267	165	40	systems	system	NOUN
cana-5267	165	41	easier	easy	ADJ
cana-5267	165	42	to	to	PART
cana-5267	165	43	deploy	deploy	VERB
cana-5267	165	44	,	,	PUNCT
cana-5267	165	45	manage	manage	VERB
cana-5267	165	46	,	,	PUNCT
cana-5267	165	47	and	and	CCONJ
cana-5267	165	48	fortify	fortify	VERB
cana-5267	165	49	.	.	PUNCT
cana-5267	166	1	algorithm	algorithm	PROPN
cana-5267	166	2	design	design	VERB
cana-5267	166	3	the	the	DET
cana-5267	166	4	below	below	ADJ
cana-5267	166	5	algorithm	algorithm	NOUN
cana-5267	166	6	outlines	outline	VERB
cana-5267	166	7	the	the	DET
cana-5267	166	8	step	step	NOUN
cana-5267	166	9	-	-	PUNCT
cana-5267	166	10	by	by	ADP
cana-5267	166	11	-	-	PUNCT
cana-5267	166	12	step	step	NOUN
cana-5267	166	13	workflow	workflow	NOUN
cana-5267	166	14	for	for	ADP
cana-5267	166	15	implementing	implement	VERB
cana-5267	166	16	automated	automate	VERB
cana-5267	166	17	threat	threat	NOUN
cana-5267	166	18	modeling	modeling	NOUN
cana-5267	166	19	across	across	ADP
cana-5267	166	20	the	the	DET
cana-5267	166	21	five	five	NUM
cana-5267	166	22	core	core	NOUN
cana-5267	166	23	modules	module	NOUN
cana-5267	166	24	:	:	PUNCT
cana-5267	166	25	data	datum	NOUN
cana-5267	166	26	sources	source	NOUN
cana-5267	166	27	,	,	PUNCT
cana-5267	166	28	threat	threat	NOUN
cana-5267	166	29	detection	detection	NOUN
cana-5267	166	30	,	,	PUNCT
cana-5267	166	31	risk	risk	NOUN
cana-5267	166	32	assessment	assessment	NOUN
cana-5267	166	33	,	,	PUNCT
cana-5267	166	34	automated	automate	VERB
cana-5267	166	35	mitigation	mitigation	NOUN
cana-5267	166	36	,	,	PUNCT
cana-5267	166	37	and	and	CCONJ
cana-5267	166	38	enterprise	enterprise	NOUN
cana-5267	166	39	integration	integration	NOUN
cana-5267	166	40	.	.	PUNCT
cana-5267	167	1	input	input	NOUN
cana-5267	167	2	:	:	PUNCT
cana-5267	167	3	training	train	VERB
cana-5267	167	4	data	datum	NOUN
cana-5267	167	5	traindb	traindb	NOUN
cana-5267	167	6	,	,	PUNCT
cana-5267	167	7	testing	test	VERB
cana-5267	167	8	data	datum	NOUN
cana-5267	167	9	tsetdb	tsetdb	PROPN
cana-5267	167	10	output	output	NOUN
cana-5267	167	11	:	:	PUNCT
cana-5267	167	12	train	train	NOUN
cana-5267	167	13	module	module	NOUN
cana-5267	167	14	tm	tm	NOUN
cana-5267	167	15	train(traindb	train(traindb	PROPN
cana-5267	167	16	)	)	PUNCT
cana-5267	167	17	step	step	NOUN
cana-5267	167	18	1	1	NUM
cana-5267	167	19	:	:	PUNCT
cana-5267	167	20	normlized	normlize	VERB
cana-5267	167	21			ADJ
cana-5267	167	22	normalization	normalization	NOUN
cana-5267	167	23	(	(	PUNCT
cana-5267	167	24	∑	∑	ADP
cana-5267	167	25	𝑇𝑟𝑎𝑖𝑛𝐷𝐵	𝑇𝑟𝑎𝑖𝑛𝐷𝐵	PROPN
cana-5267	167	26	[	[	X
cana-5267	167	27	𝑘	𝑘	X
cana-5267	167	28	]	]	X
cana-5267	167	29	.	.	PUNCT
cana-5267	168	1	𝑠𝑒𝑙𝑒𝑐𝑡	𝑠𝑒𝑙𝑒𝑐𝑡	NOUN
cana-5267	168	2	𝑖𝑓	𝑖𝑓	ADP
cana-5267	168	3	𝑑𝑒𝑠𝑐𝑟𝑒𝑡𝑒𝑛	𝑑𝑒𝑠𝑐𝑟𝑒𝑡𝑒𝑛	NOUN
cana-5267	168	4	𝑘=0	𝑘=0	PART
cana-5267	168	5	)	)	PUNCT
cana-5267	168	6	step	step	NOUN
cana-5267	168	7	2	2	NUM
cana-5267	168	8	:	:	PUNCT
cana-5267	168	9	extract	extract	VERB
cana-5267	168	10	co	co	ADJ
cana-5267	168	11	-	-	ADJ
cana-5267	168	12	relational	relational	ADJ
cana-5267	168	13	features	feature	NOUN
cana-5267	168	14	from	from	ADP
cana-5267	168	15	normalized	normalize	VERB
cana-5267	168	16	set	set	VERB
cana-5267	168	17	step	step	NOUN
cana-5267	168	18	3	3	NUM
cana-5267	168	19	:	:	PUNCT
cana-5267	168	20	i	i	PRON
cana-5267	168	21			VERB
cana-5267	168	22	0	0	NUM
cana-5267	168	23	epoch	epoch	PROPN
cana-5267	168	24			VERB
cana-5267	168	25	100	100	NUM
cana-5267	168	26	while	while	SCONJ
cana-5267	168	27	(	(	PUNCT
cana-5267	168	28	i	i	INTJ
cana-5267	168	29	!	!	PUNCT
cana-5267	168	30	=	=	PUNCT
cana-5267	169	1	epoch	epoch	NOUN
cana-5267	169	2	)	)	PUNCT
cana-5267	169	3	module	module	NOUN
cana-5267	169	4	[	[	X
cana-5267	169	5	]	]	X
cana-5267	169	6			PROPN
cana-5267	169	7	trainclassifier	trainclassifier	NOUN
cana-5267	169	8	(	(	PUNCT
cana-5267	169	9	∑	∑	PUNCT
cana-5267	169	10	𝑁𝑜𝑟𝑚𝑙𝑖𝑧𝑒𝑑	𝑁𝑜𝑟𝑚𝑙𝑖𝑧𝑒𝑑	PROPN
cana-5267	169	11	[	[	X
cana-5267	169	12	𝑘]𝑛	𝑘]𝑛	NOUN
cana-5267	169	13	𝑘=0	𝑘=0	PROPN
cana-5267	169	14	)	)	PUNCT
cana-5267	170	1	i	i	PRON
cana-5267	170	2	=	=	NOUN
cana-5267	170	3	i+1	i+1	PART
cana-5267	170	4	end	end	NOUN
cana-5267	170	5	loop	loop	NOUN
cana-5267	170	6	step	step	NOUN
cana-5267	170	7	4	4	NUM
cana-5267	170	8	:	:	PUNCT
cana-5267	170	9	return	return	VERB
cana-5267	170	10	module	module	NOUN
cana-5267	170	11	[	[	X
cana-5267	170	12	]	]	X
cana-5267	170	13	test	test	NOUN
cana-5267	170	14	(	(	PUNCT
cana-5267	170	15	testdb	testdb	PROPN
cana-5267	170	16	,	,	PUNCT
cana-5267	170	17	module	module	NOUN
cana-5267	170	18	[	[	X
cana-5267	170	19	]	]	X
cana-5267	170	20	)	)	PUNCT
cana-5267	170	21	step	step	NOUN
cana-5267	170	22	1	1	NUM
cana-5267	170	23	:	:	PUNCT
cana-5267	170	24	for	for	SCONJ
cana-5267	170	25	each	each	PRON
cana-5267	170	26	(	(	PUNCT
cana-5267	170	27	read	read	VERB
cana-5267	170	28	instance	instance	NOUN
cana-5267	170	29	from	from	ADP
cana-5267	170	30	testdb	testdb	PROPN
cana-5267	170	31	)	)	PUNCT
cana-5267	170	32	step	step	NOUN
cana-5267	170	33	2	2	NUM
cana-5267	170	34	:	:	PUNCT
cana-5267	170	35	attributesset	attributesset	VERB
cana-5267	170	36	[	[	X
cana-5267	170	37	]	]	X
cana-5267	170	38			VERB
cana-5267	170	39	instance.splite	instance.splite	ADJ
cana-5267	170	40	step	step	NOUN
cana-5267	170	41	3	3	NUM
cana-5267	170	42	:	:	PUNCT
cana-5267	170	43	calculate	calculate	VERB
cana-5267	170	44	weight	weight	NOUN
cana-5267	170	45	of	of	ADP
cana-5267	170	46	respective	respective	ADJ
cana-5267	170	47	instance	instance	NOUN
cana-5267	170	48	using	use	VERB
cana-5267	170	49	below	below	ADP
cana-5267	170	50	equation	equation	NOUN
cana-5267	170	51	𝑤𝑒𝑖𝑔ℎ𝑡	𝑤𝑒𝑖𝑔ℎ𝑡	VERB
cana-5267	170	52	=	=	SYM
cana-5267	170	53	𝑇𝑒𝑠𝑡𝑐𝑙𝑎𝑠𝑠𝑖𝑓𝑖𝑒𝑟	𝑇𝑒𝑠𝑡𝑐𝑙𝑎𝑠𝑠𝑖𝑓𝑖𝑒𝑟	PROPN
cana-5267	170	54	(	(	PUNCT
cana-5267	170	55	𝐴𝑡𝑡𝑟𝑖𝑏𝑢𝑡𝑒𝑠𝑆𝑒𝑡[𝑖𝑛𝑑𝑒𝑥	𝐴𝑡𝑡𝑟𝑖𝑏𝑢𝑡𝑒𝑠𝑆𝑒𝑡[𝑖𝑛𝑑𝑒𝑥	NUM
cana-5267	170	56	]	]	PUNCT
cana-5267	170	57	=	=	PUNCT
cana-5267	170	58	∑	∑	PUNCT
cana-5267	170	59	𝑀𝑜𝑑𝑢𝑙𝑒[𝑖	𝑀𝑜𝑑𝑢𝑙𝑒[𝑖	PROPN
cana-5267	170	60	]	]	X
cana-5267	170	61	)	)	PUNCT
cana-5267	170	62	𝑛	𝑛	PRON
cana-5267	170	63	𝑖=1	𝑖=1	PROPN
cana-5267	170	64	communications	communication	NOUN
cana-5267	170	65	on	on	ADP
cana-5267	170	66	applied	apply	VERB
cana-5267	170	67	nonlinear	nonlinear	ADJ
cana-5267	170	68	analysis	analysis	NOUN
cana-5267	170	69	issn	issn	NOUN
cana-5267	170	70	:	:	PUNCT
cana-5267	170	71	1074	1074	NUM
cana-5267	170	72	-	-	PUNCT
cana-5267	170	73	133x	133x	NUM
cana-5267	170	74	vol	vol	VERB
cana-5267	170	75	32	32	NUM
cana-5267	170	76	no	no	NOUN
cana-5267	170	77	.	.	PUNCT
cana-5267	171	1	10s	10	NOUN
cana-5267	171	2	(	(	PUNCT
cana-5267	171	3	2025	2025	NUM
cana-5267	171	4	)	)	PUNCT
cana-5267	171	5	1666	1666	NUM
cana-5267	171	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5267	171	7	step	step	NOUN
cana-5267	171	8	4	4	NUM
cana-5267	171	9	:	:	PUNCT
cana-5267	171	10	if	if	SCONJ
cana-5267	171	11	(	(	PUNCT
cana-5267	171	12	weight	weight	NOUN
cana-5267	171	13	>	>	X
cana-5267	171	14	threshold	threshold	PROPN
cana-5267	171	15	)	)	PUNCT
cana-5267	171	16	flag	flag	NOUN
cana-5267	171	17	=	=	SYM
cana-5267	171	18	1	1	NUM
cana-5267	171	19	return	return	NOUN
cana-5267	171	20	attack	attack	NOUN
cana-5267	171	21	step	step	NOUN
cana-5267	171	22	5	5	NUM
cana-5267	171	23	:	:	PUNCT
cana-5267	171	24	if(flag==0	if(flag==0	NUM
cana-5267	171	25	)	)	PUNCT
cana-5267	171	26	return	return	VERB
cana-5267	171	27	normal	normal	ADJ
cana-5267	171	28	the	the	DET
cana-5267	171	29	focus	focus	NOUN
cana-5267	171	30	of	of	ADP
cana-5267	171	31	this	this	DET
cana-5267	171	32	work	work	NOUN
cana-5267	171	33	is	be	AUX
cana-5267	171	34	formulating	formulate	VERB
cana-5267	171	35	an	an	DET
cana-5267	171	36	algorithm	algorithm	NOUN
cana-5267	171	37	for	for	ADP
cana-5267	171	38	training	train	VERB
cana-5267	171	39	a	a	DET
cana-5267	171	40	classification	classification	NOUN
cana-5267	171	41	model	model	NOUN
cana-5267	171	42	to	to	PART
cana-5267	171	43	detect	detect	VERB
cana-5267	171	44	anomalous	anomalous	ADJ
cana-5267	171	45	or	or	CCONJ
cana-5267	171	46	malicious	malicious	ADJ
cana-5267	171	47	attacks	attack	NOUN
cana-5267	171	48	using	use	VERB
cana-5267	171	49	a	a	DET
cana-5267	171	50	training	training	NOUN
cana-5267	171	51	-	-	PUNCT
cana-5267	171	52	test	test	NOUN
cana-5267	171	53	methodology	methodology	NOUN
cana-5267	171	54	tailored	tailor	VERB
cana-5267	171	55	to	to	PART
cana-5267	171	56	classify	classify	VERB
cana-5267	171	57	such	such	ADJ
cana-5267	171	58	attacks	attack	NOUN
cana-5267	171	59	.	.	PUNCT
cana-5267	172	1	the	the	DET
cana-5267	172	2	model	model	NOUN
cana-5267	172	3	input	input	NOUN
cana-5267	172	4	comprises	comprise	VERB
cana-5267	172	5	a	a	DET
cana-5267	172	6	database	database	NOUN
cana-5267	172	7	of	of	ADP
cana-5267	172	8	training	training	NOUN
cana-5267	172	9	data	datum	NOUN
cana-5267	172	10	(	(	PUNCT
cana-5267	172	11	traindb	traindb	NOUN
cana-5267	172	12	)	)	PUNCT
cana-5267	172	13	and	and	CCONJ
cana-5267	172	14	a	a	DET
cana-5267	172	15	database	database	NOUN
cana-5267	172	16	of	of	ADP
cana-5267	172	17	test	test	NOUN
cana-5267	172	18	data	datum	NOUN
cana-5267	172	19	(	(	PUNCT
cana-5267	172	20	testdb	testdb	PROPN
cana-5267	172	21	)	)	PUNCT
cana-5267	172	22	,	,	PUNCT
cana-5267	172	23	and	and	CCONJ
cana-5267	172	24	the	the	DET
cana-5267	172	25	output	output	NOUN
cana-5267	172	26	is	be	AUX
cana-5267	172	27	a	a	DET
cana-5267	172	28	trained	train	VERB
cana-5267	172	29	threat	threat	NOUN
cana-5267	172	30	detection	detection	NOUN
cana-5267	172	31	module	module	NOUN
cana-5267	172	32	(	(	PUNCT
cana-5267	172	33	tm	tm	NOUN
cana-5267	172	34	)	)	PUNCT
cana-5267	172	35	.	.	PUNCT
cana-5267	173	1	the	the	DET
cana-5267	173	2	training	training	NOUN
cana-5267	173	3	phase	phase	NOUN
cana-5267	173	4	commences	commence	VERB
cana-5267	173	5	with	with	ADP
cana-5267	173	6	normalization	normalization	NOUN
cana-5267	173	7	,	,	PUNCT
cana-5267	173	8	which	which	PRON
cana-5267	173	9	entails	entail	VERB
cana-5267	173	10	processing	process	VERB
cana-5267	173	11	the	the	DET
cana-5267	173	12	discrete	discrete	ADJ
cana-5267	173	13	values	value	NOUN
cana-5267	173	14	in	in	ADP
cana-5267	173	15	the	the	DET
cana-5267	173	16	dataset	dataset	NOUN
cana-5267	173	17	to	to	PART
cana-5267	173	18	ensure	ensure	VERB
cana-5267	173	19	uniformity	uniformity	NOUN
cana-5267	173	20	and	and	CCONJ
cana-5267	173	21	consistency	consistency	NOUN
cana-5267	173	22	.	.	PUNCT
cana-5267	174	1	to	to	PART
cana-5267	174	2	achieve	achieve	VERB
cana-5267	174	3	a	a	DET
cana-5267	174	4	high	high	ADJ
cana-5267	174	5	classification	classification	NOUN
cana-5267	174	6	accuracy	accuracy	NOUN
cana-5267	174	7	,	,	PUNCT
cana-5267	174	8	the	the	DET
cana-5267	174	9	extractor	extractor	NOUN
cana-5267	174	10	additionally	additionally	ADV
cana-5267	174	11	applies	apply	VERB
cana-5267	174	12	feature	feature	NOUN
cana-5267	174	13	selection	selection	NOUN
cana-5267	174	14	to	to	PART
cana-5267	174	15	retrieve	retrieve	VERB
cana-5267	174	16	class	class	NOUN
cana-5267	174	17	co	co	NOUN
cana-5267	174	18	-	-	ADJ
cana-5267	174	19	relational	relational	ADJ
cana-5267	174	20	features	feature	NOUN
cana-5267	174	21	pertinent	pertinent	ADJ
cana-5267	174	22	to	to	ADP
cana-5267	174	23	the	the	DET
cana-5267	174	24	task	task	NOUN
cana-5267	174	25	at	at	ADP
cana-5267	174	26	hand	hand	NOUN
cana-5267	174	27	.	.	PUNCT
cana-5267	175	1	the	the	DET
cana-5267	175	2	training	training	NOUN
cana-5267	175	3	process	process	NOUN
cana-5267	175	4	is	be	AUX
cana-5267	175	5	executed	execute	VERB
cana-5267	175	6	for	for	ADP
cana-5267	175	7	100	100	NUM
cana-5267	175	8	epochs	epoch	NOUN
cana-5267	175	9	within	within	ADP
cana-5267	175	10	which	which	PRON
cana-5267	175	11	the	the	DET
cana-5267	175	12	refinement	refinement	NOUN
cana-5267	175	13	on	on	ADP
cana-5267	175	14	the	the	DET
cana-5267	175	15	classification	classification	NOUN
cana-5267	175	16	model	model	NOUN
cana-5267	175	17	is	be	AUX
cana-5267	175	18	performed	perform	VERB
cana-5267	175	19	.	.	PUNCT
cana-5267	176	1	for	for	ADP
cana-5267	176	2	each	each	DET
cana-5267	176	3	iteration	iteration	NOUN
cana-5267	176	4	,	,	PUNCT
cana-5267	176	5	a	a	DET
cana-5267	176	6	classifier	classifier	NOUN
cana-5267	176	7	module	module	NOUN
cana-5267	176	8	is	be	AUX
cana-5267	176	9	trained	train	VERB
cana-5267	176	10	over	over	ADP
cana-5267	176	11	the	the	DET
cana-5267	176	12	normalized	normalize	VERB
cana-5267	176	13	dataset	dataset	NOUN
cana-5267	176	14	and	and	CCONJ
cana-5267	176	15	the	the	DET
cana-5267	176	16	resulting	result	VERB
cana-5267	176	17	model	model	NOUN
cana-5267	176	18	is	be	AUX
cana-5267	176	19	preserved	preserve	VERB
cana-5267	176	20	for	for	ADP
cana-5267	176	21	evaluation	evaluation	NOUN
cana-5267	176	22	.	.	PUNCT
cana-5267	177	1	after	after	ADP
cana-5267	177	2	finishing	finish	VERB
cana-5267	177	3	the	the	DET
cana-5267	177	4	training	training	NOUN
cana-5267	177	5	phase	phase	NOUN
cana-5267	177	6	,	,	PUNCT
cana-5267	177	7	all	all	DET
cana-5267	177	8	the	the	DET
cana-5267	177	9	trained	train	VERB
cana-5267	177	10	modules	module	NOUN
cana-5267	177	11	are	be	AUX
cana-5267	177	12	assembled	assemble	VERB
cana-5267	177	13	in	in	ADP
cana-5267	177	14	module	module	NOUN
cana-5267	177	15	[	[	X
cana-5267	177	16	]	]	X
cana-5267	177	17	in	in	ADP
cana-5267	177	18	anticipation	anticipation	NOUN
cana-5267	177	19	of	of	ADP
cana-5267	177	20	future	future	ADJ
cana-5267	177	21	classification	classification	NOUN
cana-5267	177	22	scenarios	scenario	NOUN
cana-5267	177	23	.	.	PUNCT
cana-5267	178	1	at	at	ADP
cana-5267	178	2	the	the	DET
cana-5267	178	3	testing	testing	NOUN
cana-5267	178	4	stage	stage	NOUN
cana-5267	178	5	,	,	PUNCT
cana-5267	178	6	every	every	DET
cana-5267	178	7	single	single	ADJ
cana-5267	178	8	new	new	ADJ
cana-5267	178	9	instance	instance	NOUN
cana-5267	178	10	from	from	ADP
cana-5267	178	11	testdb	testdb	PROPN
cana-5267	178	12	is	be	AUX
cana-5267	178	13	scrutinized	scrutinize	VERB
cana-5267	178	14	.	.	PUNCT
cana-5267	179	1	using	use	VERB
cana-5267	179	2	the	the	DET
cana-5267	179	3	trained	train	VERB
cana-5267	179	4	classification	classification	NOUN
cana-5267	179	5	modules	module	NOUN
cana-5267	179	6	,	,	PUNCT
cana-5267	179	7	the	the	DET
cana-5267	179	8	algorithm	algorithm	NOUN
cana-5267	179	9	extracts	extract	VERB
cana-5267	179	10	relevant	relevant	ADJ
cana-5267	179	11	attributes	attribute	NOUN
cana-5267	179	12	and	and	CCONJ
cana-5267	179	13	computes	compute	VERB
cana-5267	179	14	a	a	DET
cana-5267	179	15	specific	specific	ADJ
cana-5267	179	16	weight	weight	NOUN
cana-5267	179	17	.	.	PUNCT
cana-5267	180	1	should	should	AUX
cana-5267	180	2	the	the	DET
cana-5267	180	3	value	value	NOUN
cana-5267	180	4	of	of	ADP
cana-5267	180	5	the	the	DET
cana-5267	180	6	weight	weight	NOUN
cana-5267	180	7	surpass	surpass	NOUN
cana-5267	180	8	the	the	DET
cana-5267	180	9	threshold	threshold	NOUN
cana-5267	180	10	,	,	PUNCT
cana-5267	180	11	the	the	DET
cana-5267	180	12	instance	instance	NOUN
cana-5267	180	13	is	be	AUX
cana-5267	180	14	deemed	deem	VERB
cana-5267	180	15	an	an	DET
cana-5267	180	16	attack	attack	NOUN
cana-5267	180	17	;	;	PUNCT
cana-5267	180	18	conversely	conversely	ADV
cana-5267	180	19	,	,	PUNCT
cana-5267	180	20	should	should	AUX
cana-5267	180	21	it	it	PRON
cana-5267	180	22	not	not	PART
cana-5267	180	23	surpass	surpass	VERB
cana-5267	180	24	the	the	DET
cana-5267	180	25	threshold	threshold	NOUN
cana-5267	180	26	,	,	PUNCT
cana-5267	180	27	it	it	PRON
cana-5267	180	28	is	be	AUX
cana-5267	180	29	deemed	deem	VERB
cana-5267	180	30	normal	normal	ADJ
cana-5267	180	31	.	.	PUNCT
cana-5267	181	1	this	this	PRON
cana-5267	181	2	achieves	achieve	VERB
cana-5267	181	3	optimal	optimal	ADJ
cana-5267	181	4	threat	threat	NOUN
cana-5267	181	5	detection	detection	NOUN
cana-5267	181	6	thereby	thereby	ADV
cana-5267	181	7	improving	improve	VERB
cana-5267	181	8	the	the	DET
cana-5267	181	9	system	system	NOUN
cana-5267	181	10	's	's	PART
cana-5267	181	11	operational	operational	ADJ
cana-5267	181	12	integrity	integrity	NOUN
cana-5267	181	13	and	and	CCONJ
cana-5267	181	14	dependability	dependability	NOUN
cana-5267	181	15	as	as	ADV
cana-5267	181	16	well	well	ADV
cana-5267	181	17	as	as	ADP
cana-5267	181	18	enhancing	enhance	VERB
cana-5267	181	19	the	the	DET
cana-5267	181	20	overall	overall	ADJ
cana-5267	181	21	security	security	NOUN
cana-5267	181	22	framework	framework	NOUN
cana-5267	181	23	.	.	PUNCT
cana-5267	182	1	results	result	NOUN
cana-5267	182	2	and	and	CCONJ
cana-5267	182	3	discussion	discussion	NOUN
cana-5267	182	4	this	this	DET
cana-5267	182	5	work	work	NOUN
cana-5267	182	6	revolves	revolve	VERB
cana-5267	182	7	around	around	ADP
cana-5267	182	8	creating	create	VERB
cana-5267	182	9	an	an	DET
cana-5267	182	10	algorithm	algorithm	NOUN
cana-5267	182	11	capable	capable	ADJ
cana-5267	182	12	of	of	ADP
cana-5267	182	13	training	train	VERB
cana-5267	182	14	a	a	DET
cana-5267	182	15	classification	classification	NOUN
cana-5267	182	16	model	model	NOUN
cana-5267	182	17	to	to	PART
cana-5267	182	18	identify	identify	VERB
cana-5267	182	19	anomalous	anomalous	ADJ
cana-5267	182	20	or	or	CCONJ
cana-5267	182	21	malicious	malicious	ADJ
cana-5267	182	22	attacks	attack	NOUN
cana-5267	182	23	using	use	VERB
cana-5267	182	24	a	a	DET
cana-5267	182	25	customized	customized	ADJ
cana-5267	182	26	training	training	NOUN
cana-5267	182	27	-	-	PUNCT
cana-5267	182	28	test	test	NOUN
cana-5267	182	29	approach	approach	NOUN
cana-5267	182	30	meant	mean	VERB
cana-5267	182	31	for	for	ADP
cana-5267	182	32	such	such	ADJ
cana-5267	182	33	classification	classification	NOUN
cana-5267	182	34	.	.	PUNCT
cana-5267	183	1	the	the	DET
cana-5267	183	2	model	model	NOUN
cana-5267	183	3	's	's	PART
cana-5267	183	4	input	input	NOUN
cana-5267	183	5	includes	include	VERB
cana-5267	183	6	the	the	DET
cana-5267	183	7	training	training	NOUN
cana-5267	183	8	data	datum	NOUN
cana-5267	183	9	within	within	ADP
cana-5267	183	10	a	a	DET
cana-5267	183	11	database	database	NOUN
cana-5267	183	12	called	call	VERB
cana-5267	183	13	traindb	traindb	NOUN
cana-5267	183	14	,	,	PUNCT
cana-5267	183	15	a	a	DET
cana-5267	183	16	database	database	NOUN
cana-5267	183	17	of	of	ADP
cana-5267	183	18	test	test	NOUN
cana-5267	183	19	data	datum	NOUN
cana-5267	183	20	referred	refer	VERB
cana-5267	183	21	to	to	ADP
cana-5267	183	22	as	as	ADP
cana-5267	183	23	testdb	testdb	PROPN
cana-5267	183	24	,	,	PUNCT
cana-5267	183	25	and	and	CCONJ
cana-5267	183	26	the	the	DET
cana-5267	183	27	output	output	NOUN
cana-5267	183	28	is	be	AUX
cana-5267	183	29	the	the	DET
cana-5267	183	30	threat	threat	NOUN
cana-5267	183	31	detection	detection	NOUN
cana-5267	183	32	module	module	NOUN
cana-5267	183	33	,	,	PUNCT
cana-5267	183	34	denoted	denote	VERB
cana-5267	183	35	as	as	ADP
cana-5267	183	36	tm	tm	PROPN
cana-5267	183	37	,	,	PUNCT
cana-5267	183	38	which	which	PRON
cana-5267	183	39	is	be	AUX
cana-5267	183	40	undergoing	undergo	VERB
cana-5267	183	41	training	training	NOUN
cana-5267	183	42	.	.	PUNCT
cana-5267	184	1	in	in	ADP
cana-5267	184	2	the	the	DET
cana-5267	184	3	model	model	NOUN
cana-5267	184	4	's	's	PART
cana-5267	184	5	training	training	NOUN
cana-5267	184	6	phase	phase	NOUN
cana-5267	184	7	,	,	PUNCT
cana-5267	184	8	all	all	DET
cana-5267	184	9	the	the	DET
cana-5267	184	10	discrete	discrete	ADJ
cana-5267	184	11	values	value	NOUN
cana-5267	184	12	within	within	ADP
cana-5267	184	13	the	the	DET
cana-5267	184	14	dataset	dataset	NOUN
cana-5267	184	15	undergo	undergo	VERB
cana-5267	184	16	a	a	DET
cana-5267	184	17	uniformity	uniformity	NOUN
cana-5267	184	18	and	and	CCONJ
cana-5267	184	19	consistency	consistency	NOUN
cana-5267	184	20	procedure	procedure	NOUN
cana-5267	184	21	termed	term	VERB
cana-5267	184	22	normalization	normalization	NOUN
cana-5267	184	23	,	,	PUNCT
cana-5267	184	24	or	or	CCONJ
cana-5267	184	25	more	more	ADV
cana-5267	184	26	colloquially	colloquially	ADV
cana-5267	184	27	,	,	PUNCT
cana-5267	184	28	“	"	PUNCT
cana-5267	184	29	value	value	NOUN
cana-5267	184	30	adjusting	adjust	VERB
cana-5267	184	31	.	.	PUNCT
cana-5267	184	32	”	"	PUNCT
cana-5267	185	1	in	in	ADP
cana-5267	185	2	order	order	NOUN
cana-5267	185	3	to	to	PART
cana-5267	185	4	achieve	achieve	VERB
cana-5267	185	5	a	a	DET
cana-5267	185	6	refined	refined	ADJ
cana-5267	185	7	classification	classification	NOUN
cana-5267	185	8	accuracy	accuracy	NOUN
cana-5267	185	9	,	,	PUNCT
cana-5267	185	10	the	the	DET
cana-5267	185	11	extractor	extractor	NOUN
cana-5267	185	12	also	also	ADV
cana-5267	185	13	conducts	conduct	VERB
cana-5267	185	14	feature	feature	NOUN
cana-5267	185	15	selection	selection	NOUN
cana-5267	185	16	in	in	ADP
cana-5267	185	17	order	order	NOUN
cana-5267	185	18	to	to	PART
cana-5267	185	19	obtain	obtain	VERB
cana-5267	185	20	class	class	NOUN
cana-5267	185	21	correlating	correlate	VERB
cana-5267	185	22	features	feature	NOUN
cana-5267	185	23	necessary	necessary	ADJ
cana-5267	185	24	for	for	ADP
cana-5267	185	25	the	the	DET
cana-5267	185	26	assignment	assignment	NOUN
cana-5267	185	27	.	.	PUNCT
cana-5267	186	1	the	the	DET
cana-5267	186	2	training	training	NOUN
cana-5267	186	3	comprises	comprise	NOUN
cana-5267	186	4	of	of	ADP
cana-5267	186	5	100	100	NUM
cana-5267	186	6	epochs	epoch	NOUN
cana-5267	186	7	during	during	ADP
cana-5267	186	8	which	which	PRON
cana-5267	186	9	the	the	DET
cana-5267	186	10	iteration	iteration	NOUN
cana-5267	186	11	improves	improve	VERB
cana-5267	186	12	the	the	DET
cana-5267	186	13	classification	classification	NOUN
cana-5267	186	14	model	model	NOUN
cana-5267	186	15	.	.	PUNCT
cana-5267	187	1	a	a	DET
cana-5267	187	2	classifier	classifier	NOUN
cana-5267	187	3	module	module	NOUN
cana-5267	187	4	is	be	AUX
cana-5267	187	5	first	first	ADV
cana-5267	187	6	trained	train	VERB
cana-5267	187	7	over	over	ADP
cana-5267	187	8	the	the	DET
cana-5267	187	9	normalized	normalize	VERB
cana-5267	187	10	dataset	dataset	NOUN
cana-5267	187	11	,	,	PUNCT
cana-5267	187	12	then	then	ADV
cana-5267	187	13	the	the	DET
cana-5267	187	14	resulting	result	VERB
cana-5267	187	15	model	model	NOUN
cana-5267	187	16	saved	save	VERB
cana-5267	187	17	for	for	ADP
cana-5267	187	18	evaluation	evaluation	NOUN
cana-5267	187	19	,	,	PUNCT
cana-5267	187	20	and	and	CCONJ
cana-5267	187	21	when	when	SCONJ
cana-5267	187	22	the	the	DET
cana-5267	187	23	training	training	NOUN
cana-5267	187	24	phase	phase	NOUN
cana-5267	187	25	is	be	AUX
cana-5267	187	26	complete	complete	ADJ
cana-5267	187	27	,	,	PUNCT
cana-5267	187	28	every	every	DET
cana-5267	187	29	module	module	NOUN
cana-5267	187	30	that	that	PRON
cana-5267	187	31	has	have	AUX
cana-5267	187	32	been	be	AUX
cana-5267	187	33	trained	train	VERB
cana-5267	187	34	gets	get	AUX
cana-5267	187	35	stored	store	VERB
cana-5267	187	36	in	in	ADP
cana-5267	187	37	anticipation	anticipation	NOUN
cana-5267	187	38	of	of	ADP
cana-5267	187	39	every	every	DET
cana-5267	187	40	possible	possible	ADJ
cana-5267	187	41	future	future	ADJ
cana-5267	187	42	classification	classification	NOUN
cana-5267	187	43	scenario	scenario	NOUN
cana-5267	187	44	in	in	ADP
cana-5267	187	45	module	module	NOUN
cana-5267	187	46	[	[	X
cana-5267	187	47	]	]	X
cana-5267	187	48	.	.	PUNCT
cana-5267	188	1	during	during	ADP
cana-5267	188	2	testing	testing	NOUN
cana-5267	188	3	,	,	PUNCT
cana-5267	188	4	every	every	DET
cana-5267	188	5	new	new	ADJ
cana-5267	188	6	instance	instance	NOUN
cana-5267	188	7	of	of	ADP
cana-5267	188	8	testdb	testdb	PROPN
cana-5267	188	9	undergoes	undergo	VERB
cana-5267	188	10	scrutiny	scrutiny	NOUN
cana-5267	188	11	,	,	PUNCT
cana-5267	188	12	during	during	ADP
cana-5267	188	13	which	which	PRON
cana-5267	188	14	the	the	DET
cana-5267	188	15	algorithm	algorithm	NOUN
cana-5267	188	16	uses	use	VERB
cana-5267	188	17	the	the	DET
cana-5267	188	18	modules	module	NOUN
cana-5267	188	19	with	with	ADP
cana-5267	188	20	pre	pre	ADJ
cana-5267	188	21	-	-	ADJ
cana-5267	188	22	trained	train	VERB
cana-5267	188	23	classes	class	NOUN
cana-5267	188	24	to	to	PART
cana-5267	188	25	extract	extract	VERB
cana-5267	188	26	the	the	DET
cana-5267	188	27	needed	need	VERB
cana-5267	188	28	attributes	attribute	NOUN
cana-5267	188	29	and	and	CCONJ
cana-5267	188	30	calculate	calculate	VERB
cana-5267	188	31	a	a	DET
cana-5267	188	32	certain	certain	ADJ
cana-5267	188	33	weight	weight	NOUN
cana-5267	188	34	associated	associate	VERB
cana-5267	188	35	with	with	ADP
cana-5267	188	36	them	they	PRON
cana-5267	188	37	.	.	PUNCT
cana-5267	189	1	if	if	SCONJ
cana-5267	189	2	the	the	DET
cana-5267	189	3	weight	weight	NOUN
cana-5267	189	4	value	value	NOUN
cana-5267	189	5	exceeds	exceed	VERB
cana-5267	189	6	a	a	DET
cana-5267	189	7	certain	certain	ADJ
cana-5267	189	8	threshold	threshold	NOUN
cana-5267	189	9	,	,	PUNCT
cana-5267	189	10	the	the	DET
cana-5267	189	11	instance	instance	NOUN
cana-5267	189	12	is	be	AUX
cana-5267	189	13	marked	mark	VERB
cana-5267	189	14	as	as	ADP
cana-5267	189	15	an	an	DET
cana-5267	189	16	attack	attack	NOUN
cana-5267	189	17	;	;	PUNCT
cana-5267	189	18	if	if	SCONJ
cana-5267	189	19	it	it	PRON
cana-5267	189	20	does	do	AUX
cana-5267	189	21	n’t	not	PART
cana-5267	189	22	surpass	surpass	VERB
cana-5267	189	23	the	the	DET
cana-5267	189	24	threshold	threshold	NOUN
cana-5267	189	25	,	,	PUNCT
cana-5267	189	26	it	it	PRON
cana-5267	189	27	is	be	AUX
cana-5267	189	28	categorized	categorize	VERB
cana-5267	189	29	as	as	ADP
cana-5267	189	30	normal	normal	ADJ
cana-5267	189	31	.	.	PUNCT
cana-5267	190	1	this	this	DET
cana-5267	190	2	aids	aid	NOUN
cana-5267	190	3	in	in	ADP
cana-5267	190	4	maximizing	maximize	VERB
cana-5267	190	5	threat	threat	NOUN
cana-5267	190	6	detection	detection	NOUN
cana-5267	190	7	which	which	PRON
cana-5267	190	8	in	in	ADP
cana-5267	190	9	turn	turn	NOUN
cana-5267	190	10	increases	increase	VERB
cana-5267	190	11	the	the	DET
cana-5267	190	12	functionality	functionality	NOUN
cana-5267	190	13	and	and	CCONJ
cana-5267	190	14	reliability	reliability	NOUN
cana-5267	190	15	of	of	ADP
cana-5267	190	16	the	the	DET
cana-5267	190	17	system	system	NOUN
cana-5267	190	18	as	as	ADV
cana-5267	190	19	well	well	ADV
cana-5267	190	20	as	as	ADP
cana-5267	190	21	its	its	PRON
cana-5267	190	22	security	security	NOUN
cana-5267	190	23	posture	posture	NOUN
cana-5267	190	24	.	.	PUNCT
cana-5267	191	1	communications	communication	NOUN
cana-5267	191	2	on	on	ADP
cana-5267	191	3	applied	apply	VERB
cana-5267	191	4	nonlinear	nonlinear	ADJ
cana-5267	191	5	analysis	analysis	NOUN
cana-5267	191	6	issn	issn	NOUN
cana-5267	191	7	:	:	PUNCT
cana-5267	191	8	1074	1074	NUM
cana-5267	191	9	-	-	PUNCT
cana-5267	191	10	133x	133x	NUM
cana-5267	191	11	vol	vol	VERB
cana-5267	191	12	32	32	NUM
cana-5267	191	13	no	no	NOUN
cana-5267	191	14	.	.	PUNCT
cana-5267	192	1	10s	10	NOUN
cana-5267	192	2	(	(	PUNCT
cana-5267	192	3	2025	2025	NUM
cana-5267	192	4	)	)	PUNCT
cana-5267	192	5	1667	1667	NUM
cana-5267	192	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5267	192	7	table	table	NOUN
cana-5267	192	8	1	1	NUM
cana-5267	192	9	:	:	PUNCT
cana-5267	192	10	evaluation	evaluation	NOUN
cana-5267	192	11	of	of	ADP
cana-5267	192	12	proposed	propose	VERB
cana-5267	192	13	model	model	NOUN
cana-5267	192	14	module	module	NOUN
cana-5267	192	15	detection	detection	NOUN
cana-5267	192	16	accuracy	accuracy	NOUN
cana-5267	192	17	(	(	PUNCT
cana-5267	192	18	%	%	INTJ
cana-5267	192	19	)	)	PUNCT
cana-5267	192	20	false	false	ADJ
cana-5267	192	21	positive	positive	ADJ
cana-5267	192	22	rate	rate	NOUN
cana-5267	192	23	(	(	PUNCT
cana-5267	192	24	%	%	INTJ
cana-5267	192	25	)	)	PUNCT
cana-5267	192	26	response	response	NOUN
cana-5267	192	27	time	time	NOUN
cana-5267	192	28	(	(	PUNCT
cana-5267	192	29	seconds	second	NOUN
cana-5267	192	30	)	)	PUNCT
cana-5267	192	31	improvement	improvement	NOUN
cana-5267	192	32	over	over	ADP
cana-5267	192	33	traditional	traditional	ADJ
cana-5267	192	34	(	(	PUNCT
cana-5267	192	35	%	%	NOUN
cana-5267	192	36	)	)	PUNCT
cana-5267	192	37	data	datum	NOUN
cana-5267	192	38	sources	source	NOUN
cana-5267	192	39	92.5	92.5	NUM
cana-5267	192	40	8.5	8.5	NUM
cana-5267	192	41	2.3	2.3	NUM
cana-5267	192	42	85	85	NUM
cana-5267	192	43	threat	threat	NOUN
cana-5267	192	44	detection	detection	NOUN
cana-5267	192	45	96.8	96.8	NUM
cana-5267	192	46	7.5	7.5	NUM
cana-5267	192	47	2.0	2.0	NUM
cana-5267	192	48	87.3	87.3	NUM
cana-5267	192	49	risk	risk	NOUN
cana-5267	192	50	assessment	assessment	NOUN
cana-5267	192	51	88.4	88.4	NUM
cana-5267	192	52	11.6	11.6	NUM
cana-5267	192	53	3.2	3.2	NUM
cana-5267	192	54	23	23	NUM
cana-5267	192	55	automated	automate	VERB
cana-5267	192	56	mitigation	mitigation	NOUN
cana-5267	192	57	98.2	98.2	NUM
cana-5267	192	58	5.8	5.8	NUM
cana-5267	192	59	1.8	1.8	NUM
cana-5267	192	60	91	91	NUM
cana-5267	192	61	enterprise	enterprise	NOUN
cana-5267	192	62	integration	integration	NOUN
cana-5267	192	63	94.6	94.6	NUM
cana-5267	192	64	6.4	6.4	NUM
cana-5267	192	65	1.95	1.95	NUM
cana-5267	192	66	38	38	NUM
cana-5267	192	67	the	the	DET
cana-5267	192	68	results	result	NOUN
cana-5267	192	69	confirm	confirm	VERB
cana-5267	192	70	that	that	SCONJ
cana-5267	192	71	automated	automate	VERB
cana-5267	192	72	threat	threat	NOUN
cana-5267	192	73	modeling	modeling	NOUN
cana-5267	192	74	significantly	significantly	ADV
cana-5267	192	75	improves	improve	VERB
cana-5267	192	76	llm	llm	PROPN
cana-5267	192	77	system	system	NOUN
cana-5267	192	78	security	security	NOUN
cana-5267	192	79	by	by	ADP
cana-5267	192	80	enhancing	enhance	VERB
cana-5267	192	81	threat	threat	NOUN
cana-5267	192	82	detection	detection	NOUN
cana-5267	192	83	,	,	PUNCT
cana-5267	192	84	dynamic	dynamic	ADJ
cana-5267	192	85	risk	risk	NOUN
cana-5267	192	86	assessment	assessment	NOUN
cana-5267	192	87	,	,	PUNCT
cana-5267	192	88	and	and	CCONJ
cana-5267	192	89	real	real	ADJ
cana-5267	192	90	-	-	PUNCT
cana-5267	192	91	time	time	NOUN
cana-5267	192	92	mitigation	mitigation	NOUN
cana-5267	192	93	.	.	PUNCT
cana-5267	193	1	the	the	DET
cana-5267	193	2	modular	modular	ADJ
cana-5267	193	3	architecture	architecture	NOUN
cana-5267	193	4	ensures	ensure	VERB
cana-5267	193	5	scalability	scalability	NOUN
cana-5267	193	6	and	and	CCONJ
cana-5267	193	7	adaptability	adaptability	NOUN
cana-5267	193	8	,	,	PUNCT
cana-5267	193	9	allowing	allow	VERB
cana-5267	193	10	integration	integration	NOUN
cana-5267	193	11	with	with	ADP
cana-5267	193	12	enterprise	enterprise	NOUN
cana-5267	193	13	cybersecurity	cybersecurity	NOUN
cana-5267	193	14	ecosystems	ecosystem	NOUN
cana-5267	193	15	.	.	PUNCT
cana-5267	194	1	key	key	ADJ
cana-5267	194	2	advantages	advantage	NOUN
cana-5267	194	3	of	of	ADP
cana-5267	194	4	the	the	DET
cana-5267	194	5	proposed	propose	VERB
cana-5267	194	6	system	system	NOUN
cana-5267	194	7	.	.	PUNCT
cana-5267	195	1	despite	despite	SCONJ
cana-5267	195	2	its	its	PRON
cana-5267	195	3	effectiveness	effectiveness	NOUN
cana-5267	195	4	,	,	PUNCT
cana-5267	195	5	challenges	challenge	NOUN
cana-5267	195	6	remain	remain	VERB
cana-5267	195	7	in	in	ADP
cana-5267	195	8	handling	handle	VERB
cana-5267	195	9	zero	zero	NUM
cana-5267	195	10	-	-	PUNCT
cana-5267	195	11	day	day	NOUN
cana-5267	195	12	llm	llm	ADJ
cana-5267	195	13	-	-	ADJ
cana-5267	195	14	specific	specific	ADJ
cana-5267	195	15	attacks	attack	NOUN
cana-5267	195	16	,	,	PUNCT
cana-5267	195	17	ensuring	ensure	VERB
cana-5267	195	18	model	model	NOUN
cana-5267	195	19	explainability	explainability	NOUN
cana-5267	195	20	for	for	ADP
cana-5267	195	21	non	non	ADJ
cana-5267	195	22	-	-	ADJ
cana-5267	195	23	technical	technical	ADJ
cana-5267	195	24	security	security	NOUN
cana-5267	195	25	teams	team	NOUN
cana-5267	195	26	,	,	PUNCT
cana-5267	195	27	and	and	CCONJ
cana-5267	195	28	mitigating	mitigate	VERB
cana-5267	195	29	evolving	evolve	VERB
cana-5267	195	30	adversarial	adversarial	ADJ
cana-5267	195	31	ai	ai	NOUN
cana-5267	195	32	threats	threat	NOUN
cana-5267	195	33	.	.	PUNCT
cana-5267	196	1	figure	figure	VERB
cana-5267	196	2	1	1	NUM
cana-5267	196	3	:	:	PUNCT
cana-5267	196	4	detection	detection	NOUN
cana-5267	196	5	accuracy	accuracy	NOUN
cana-5267	196	6	analysis	analysis	NOUN
cana-5267	196	7	of	of	ADP
cana-5267	196	8	various	various	ADJ
cana-5267	196	9	modules	module	NOUN
cana-5267	196	10	communications	communication	NOUN
cana-5267	196	11	on	on	ADP
cana-5267	196	12	applied	apply	VERB
cana-5267	196	13	nonlinear	nonlinear	ADJ
cana-5267	196	14	analysis	analysis	NOUN
cana-5267	196	15	issn	issn	NOUN
cana-5267	196	16	:	:	PUNCT
cana-5267	196	17	1074	1074	NUM
cana-5267	196	18	-	-	PUNCT
cana-5267	196	19	133x	133x	NUM
cana-5267	196	20	vol	vol	VERB
cana-5267	196	21	32	32	NUM
cana-5267	196	22	no	no	NOUN
cana-5267	196	23	.	.	PUNCT
cana-5267	197	1	10s	10	NOUN
cana-5267	197	2	(	(	PUNCT
cana-5267	197	3	2025	2025	NUM
cana-5267	197	4	)	)	PUNCT
cana-5267	197	5	1668	1668	NUM
cana-5267	197	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5267	198	1	in	in	ADP
cana-5267	198	2	figure	figure	NOUN
cana-5267	198	3	2	2	NUM
cana-5267	198	4	the	the	DET
cana-5267	198	5	automated	automate	VERB
cana-5267	198	6	mitigation	mitigation	NOUN
cana-5267	198	7	module	module	NOUN
cana-5267	198	8	achieved	achieve	VERB
cana-5267	198	9	the	the	DET
cana-5267	198	10	highest	high	ADJ
cana-5267	198	11	accuracy	accuracy	NOUN
cana-5267	198	12	(	(	PUNCT
cana-5267	198	13	98.2	98.2	NUM
cana-5267	198	14	%	%	NOUN
cana-5267	198	15	)	)	PUNCT
cana-5267	198	16	,	,	PUNCT
cana-5267	198	17	followed	follow	VERB
cana-5267	198	18	by	by	ADP
cana-5267	198	19	threat	threat	NOUN
cana-5267	198	20	detection	detection	NOUN
cana-5267	198	21	(	(	PUNCT
cana-5267	198	22	96.8	96.8	NUM
cana-5267	198	23	%	%	NOUN
cana-5267	198	24	)	)	PUNCT
cana-5267	198	25	and	and	CCONJ
cana-5267	198	26	enterprise	enterprise	NOUN
cana-5267	198	27	integration	integration	NOUN
cana-5267	198	28	(	(	PUNCT
cana-5267	198	29	94.6	94.6	NUM
cana-5267	198	30	%	%	NOUN
cana-5267	198	31	)	)	PUNCT
cana-5267	198	32	.	.	PUNCT
cana-5267	199	1	figure	figure	VERB
cana-5267	199	2	3	3	NUM
cana-5267	199	3	:	:	PUNCT
cana-5267	199	4	fpr	fpr	NOUN
cana-5267	199	5	rate	rate	NOUN
cana-5267	199	6	for	for	ADP
cana-5267	199	7	different	different	ADJ
cana-5267	199	8	detection	detection	NOUN
cana-5267	199	9	modules	module	NOUN
cana-5267	199	10	in	in	ADP
cana-5267	199	11	figure	figure	NOUN
cana-5267	199	12	3	3	NUM
cana-5267	199	13	show	show	VERB
cana-5267	199	14	false	false	ADJ
cana-5267	199	15	positive	positive	ADJ
cana-5267	199	16	rate	rate	NOUN
cana-5267	199	17	(	(	PUNCT
cana-5267	199	18	fpr	fpr	NOUN
cana-5267	199	19	)	)	PUNCT
cana-5267	199	20	.	.	PUNCT
cana-5267	200	1	automated	automate	VERB
cana-5267	200	2	mitigation	mitigation	NOUN
cana-5267	200	3	(	(	PUNCT
cana-5267	200	4	5.8	5.8	NUM
cana-5267	200	5	%	%	NOUN
cana-5267	200	6	)	)	PUNCT
cana-5267	200	7	had	have	VERB
cana-5267	200	8	the	the	DET
cana-5267	200	9	lowest	low	ADJ
cana-5267	200	10	false	false	ADJ
cana-5267	200	11	positives	positive	NOUN
cana-5267	200	12	,	,	PUNCT
cana-5267	200	13	demonstrating	demonstrate	VERB
cana-5267	200	14	its	its	PRON
cana-5267	200	15	efficiency	efficiency	NOUN
cana-5267	200	16	in	in	ADP
cana-5267	200	17	blocking	block	VERB
cana-5267	200	18	adversarial	adversarial	ADJ
cana-5267	200	19	attacks	attack	NOUN
cana-5267	200	20	.	.	PUNCT
cana-5267	201	1	figure	figure	VERB
cana-5267	201	2	4	4	NUM
cana-5267	201	3	:	:	PUNCT
cana-5267	201	4	analysis	analysis	NOUN
cana-5267	201	5	of	of	ADP
cana-5267	201	6	various	various	ADJ
cana-5267	201	7	methods	method	NOUN
cana-5267	201	8	proposed	propose	VERB
cana-5267	201	9	vs	vs	ADP
cana-5267	201	10	existing	exist	VERB
cana-5267	201	11	communications	communication	NOUN
cana-5267	201	12	on	on	ADP
cana-5267	201	13	applied	apply	VERB
cana-5267	201	14	nonlinear	nonlinear	ADJ
cana-5267	201	15	analysis	analysis	NOUN
cana-5267	201	16	issn	issn	NOUN
cana-5267	201	17	:	:	PUNCT
cana-5267	201	18	1074	1074	NUM
cana-5267	201	19	-	-	PUNCT
cana-5267	201	20	133x	133x	NUM
cana-5267	201	21	vol	vol	VERB
cana-5267	201	22	32	32	NUM
cana-5267	201	23	no	no	NOUN
cana-5267	201	24	.	.	PUNCT
cana-5267	202	1	10s	10	NOUN
cana-5267	202	2	(	(	PUNCT
cana-5267	202	3	2025	2025	NUM
cana-5267	202	4	)	)	PUNCT
cana-5267	202	5	1669	1669	NUM
cana-5267	202	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5267	202	7	as	as	SCONJ
cana-5267	202	8	illustrated	illustrate	VERB
cana-5267	202	9	in	in	ADP
cana-5267	202	10	figure	figure	NOUN
cana-5267	202	11	4	4	NUM
cana-5267	202	12	,	,	PUNCT
cana-5267	202	13	the	the	DET
cana-5267	202	14	automated	automate	VERB
cana-5267	202	15	mitigation	mitigation	NOUN
cana-5267	202	16	module	module	NOUN
cana-5267	202	17	increased	increase	VERB
cana-5267	202	18	security	security	NOUN
cana-5267	202	19	protocols	protocol	NOUN
cana-5267	202	20	by	by	ADP
cana-5267	202	21	91	91	NUM
cana-5267	202	22	%	%	NOUN
cana-5267	202	23	while	while	SCONJ
cana-5267	202	24	the	the	DET
cana-5267	202	25	enterprise	enterprise	NOUN
cana-5267	202	26	integration	integration	NOUN
cana-5267	202	27	layer	layer	NOUN
cana-5267	202	28	increased	increase	VERB
cana-5267	202	29	threat	threat	NOUN
cana-5267	202	30	response	response	NOUN
cana-5267	202	31	by	by	ADP
cana-5267	202	32	38	38	NUM
cana-5267	202	33	%	%	NOUN
cana-5267	202	34	.	.	PUNCT
cana-5267	203	1	both	both	DET
cana-5267	203	2	improvements	improvement	NOUN
cana-5267	203	3	were	be	AUX
cana-5267	203	4	made	make	VERB
cana-5267	203	5	when	when	SCONJ
cana-5267	203	6	compared	compare	VERB
cana-5267	203	7	to	to	ADP
cana-5267	203	8	traditional	traditional	ADJ
cana-5267	203	9	security	security	NOUN
cana-5267	203	10	frameworks	framework	NOUN
cana-5267	203	11	the	the	DET
cana-5267	203	12	later	later	ADV
cana-5267	203	13	outlined	outline	VERB
cana-5267	203	14	plans	plan	NOUN
cana-5267	203	15	of	of	ADP
cana-5267	203	16	our	our	PRON
cana-5267	203	17	project	project	NOUN
cana-5267	203	18	include	include	VERB
cana-5267	203	19	:	:	PUNCT
cana-5267	203	20	integrating	integrate	VERB
cana-5267	203	21	advanced	advanced	ADJ
cana-5267	203	22	adversarial	adversarial	ADJ
cana-5267	203	23	training	training	NOUN
cana-5267	203	24	,	,	PUNCT
cana-5267	203	25	federated	federate	VERB
cana-5267	203	26	learning	learning	NOUN
cana-5267	203	27	to	to	PART
cana-5267	203	28	enhance	enhance	VERB
cana-5267	203	29	distributed	distribute	VERB
cana-5267	203	30	security	security	NOUN
cana-5267	203	31	intelligence	intelligence	NOUN
cana-5267	203	32	,	,	PUNCT
cana-5267	203	33	and	and	CCONJ
cana-5267	203	34	applying	apply	VERB
cana-5267	203	35	ai	ai	NOUN
cana-5267	203	36	for	for	ADP
cana-5267	203	37	policy	policy	NOUN
cana-5267	203	38	adjustment	adjustment	NOUN
cana-5267	203	39	to	to	PART
cana-5267	203	40	strengthen	strengthen	VERB
cana-5267	203	41	llm	llm	PROPN
cana-5267	203	42	cybersecurity	cybersecurity	NOUN
cana-5267	203	43	frameworks	framework	NOUN
cana-5267	203	44	even	even	ADV
cana-5267	203	45	more	more	ADV
cana-5267	203	46	.	.	PUNCT
cana-5267	204	1	conclusion	conclusion	VERB
cana-5267	204	2	the	the	DET
cana-5267	204	3	automated	automate	VERB
cana-5267	204	4	threat	threat	NOUN
cana-5267	204	5	modeling	model	VERB
cana-5267	204	6	system	system	NOUN
cana-5267	204	7	applies	apply	VERB
cana-5267	205	1	machine	machine	NOUN
cana-5267	205	2	learning	learn	VERB
cana-5267	205	3	for	for	ADP
cana-5267	205	4	threat	threat	NOUN
cana-5267	205	5	detection	detection	NOUN
cana-5267	205	6	,	,	PUNCT
cana-5267	205	7	risk	risk	NOUN
cana-5267	205	8	assessment	assessment	NOUN
cana-5267	205	9	,	,	PUNCT
cana-5267	205	10	and	and	CCONJ
cana-5267	205	11	real	real	ADJ
cana-5267	205	12	-	-	PUNCT
cana-5267	205	13	time	time	NOUN
cana-5267	205	14	mitigation	mitigation	NOUN
cana-5267	205	15	,	,	PUNCT
cana-5267	205	16	increasing	increase	VERB
cana-5267	205	17	the	the	DET
cana-5267	205	18	security	security	NOUN
cana-5267	205	19	and	and	CCONJ
cana-5267	205	20	resilience	resilience	NOUN
cana-5267	205	21	of	of	ADP
cana-5267	205	22	llm	llm	PROPN
cana-5267	205	23	systems	system	NOUN
cana-5267	205	24	.	.	PUNCT
cana-5267	206	1	results	result	NOUN
cana-5267	206	2	show	show	VERB
cana-5267	206	3	that	that	SCONJ
cana-5267	206	4	the	the	DET
cana-5267	206	5	system	system	NOUN
cana-5267	206	6	is	be	AUX
cana-5267	206	7	able	able	ADJ
cana-5267	206	8	to	to	PART
cana-5267	206	9	accurately	accurately	ADV
cana-5267	206	10	detect	detect	VERB
cana-5267	206	11	adversarial	adversarial	ADJ
cana-5267	206	12	threats	threat	NOUN
cana-5267	206	13	at	at	ADP
cana-5267	206	14	96.8	96.8	NUM
cana-5267	206	15	%	%	NOUN
cana-5267	206	16	accuracy	accuracy	NOUN
cana-5267	206	17	,	,	PUNCT
cana-5267	206	18	lowering	lower	VERB
cana-5267	206	19	the	the	DET
cana-5267	206	20	misclassification	misclassification	NOUN
cana-5267	206	21	rate	rate	NOUN
cana-5267	206	22	by	by	ADP
cana-5267	206	23	23	23	NUM
cana-5267	206	24	%	%	NOUN
cana-5267	206	25	and	and	CCONJ
cana-5267	206	26	improving	improve	VERB
cana-5267	206	27	response	response	NOUN
cana-5267	206	28	time	time	NOUN
cana-5267	206	29	over	over	ADP
cana-5267	206	30	traditional	traditional	ADJ
cana-5267	206	31	methods	method	NOUN
cana-5267	206	32	.	.	PUNCT
cana-5267	207	1	the	the	DET
cana-5267	207	2	system	system	NOUN
cana-5267	207	3	also	also	ADV
cana-5267	207	4	enhances	enhance	VERB
cana-5267	207	5	security	security	NOUN
cana-5267	207	6	automation	automation	NOUN
cana-5267	207	7	with	with	ADP
cana-5267	207	8	ease	ease	NOUN
cana-5267	207	9	of	of	ADP
cana-5267	207	10	integration	integration	NOUN
cana-5267	207	11	to	to	ADP
cana-5267	207	12	siem	siem	PROPN
cana-5267	207	13	and	and	CCONJ
cana-5267	207	14	edr	edr	PROPN
cana-5267	207	15	systems	system	NOUN
cana-5267	207	16	.	.	PUNCT
cana-5267	208	1	although	although	SCONJ
cana-5267	208	2	there	there	PRON
cana-5267	208	3	are	be	VERB
cana-5267	208	4	still	still	ADV
cana-5267	208	5	some	some	DET
cana-5267	208	6	challenges	challenge	NOUN
cana-5267	208	7	in	in	ADP
cana-5267	208	8	addressing	address	VERB
cana-5267	208	9	zero	zero	NUM
cana-5267	208	10	-	-	PUNCT
cana-5267	208	11	day	day	NOUN
cana-5267	208	12	attacks	attack	NOUN
cana-5267	208	13	,	,	PUNCT
cana-5267	208	14	future	future	ADJ
cana-5267	208	15	work	work	NOUN
cana-5267	208	16	in	in	ADP
cana-5267	208	17	adversarial	adversarial	ADJ
cana-5267	208	18	training	training	NOUN
cana-5267	208	19	and	and	CCONJ
cana-5267	208	20	ai	ai	VERB
cana-5267	208	21	policy	policy	NOUN
cana-5267	208	22	modification	modification	NOUN
cana-5267	208	23	will	will	AUX
cana-5267	208	24	strengthen	strengthen	VERB
cana-5267	208	25	llm	llm	PROPN
cana-5267	208	26	cybersecurity	cybersecurity	NOUN
cana-5267	208	27	architectures	architecture	NOUN
cana-5267	208	28	,	,	PUNCT
cana-5267	208	29	providing	provide	VERB
cana-5267	208	30	scalable	scalable	ADJ
cana-5267	208	31	and	and	CCONJ
cana-5267	208	32	proactive	proactive	ADJ
cana-5267	208	33	defense	defense	NOUN
cana-5267	208	34	strategies	strategy	NOUN
cana-5267	208	35	.	.	PUNCT
cana-5267	209	1	references	reference	NOUN
cana-5267	209	2	[	[	X
cana-5267	209	3	1	1	NUM
cana-5267	209	4	.	.	PUNCT
cana-5267	209	5	]	]	PUNCT
cana-5267	210	1	brown	brown	ADJ
cana-5267	210	2	,	,	PUNCT
cana-5267	210	3	n.	n.	NOUN
cana-5267	210	4	,	,	PUNCT
cana-5267	210	5	&	&	CCONJ
cana-5267	210	6	johnson	johnson	PROPN
cana-5267	210	7	,	,	PUNCT
cana-5267	210	8	m.	m.	NOUN
cana-5267	210	9	(	(	PUNCT
cana-5267	210	10	2023	2023	NUM
cana-5267	210	11	)	)	PUNCT
cana-5267	210	12	.	.	PUNCT
cana-5267	211	1	securing	secure	VERB
cana-5267	211	2	large	large	ADJ
cana-5267	211	3	language	language	NOUN
cana-5267	211	4	models	model	NOUN
cana-5267	211	5	:	:	PUNCT
cana-5267	211	6	an	an	DET
cana-5267	211	7	automated	automate	VERB
cana-5267	211	8	threat	threat	NOUN
cana-5267	211	9	modeling	modeling	NOUN
cana-5267	211	10	approach	approach	NOUN
cana-5267	211	11	.	.	PUNCT
cana-5267	212	1	journal	journal	NOUN
cana-5267	212	2	of	of	ADP
cana-5267	212	3	ai	ai	PROPN
cana-5267	212	4	security	security	NOUN
cana-5267	212	5	,	,	PUNCT
cana-5267	212	6	9(2	9(2	NUM
cana-5267	212	7	)	)	PUNCT
cana-5267	212	8	,	,	PUNCT
cana-5267	212	9	112	112	NUM
cana-5267	212	10	-	-	SYM
cana-5267	212	11	134	134	NUM
cana-5267	212	12	.	.	PUNCT
cana-5267	213	1	https://doi.org/10.1016/j.aisec.2023.002	https://doi.org/10.1016/j.aisec.2023.002	NOUN
cana-5267	214	1	[	[	X
cana-5267	214	2	2	2	NUM
cana-5267	214	3	.	.	PUNCT
cana-5267	214	4	]	]	X
cana-5267	215	1	chen	chen	PROPN
cana-5267	215	2	,	,	PUNCT
cana-5267	215	3	l.	l.	PROPN
cana-5267	215	4	,	,	PUNCT
cana-5267	215	5	wang	wang	PROPN
cana-5267	215	6	,	,	PUNCT
cana-5267	215	7	h.	h.	PROPN
cana-5267	215	8	,	,	PUNCT
cana-5267	215	9	&	&	CCONJ
cana-5267	215	10	liu	liu	PROPN
cana-5267	215	11	,	,	PUNCT
cana-5267	215	12	x.	x.	NOUN
cana-5267	215	13	(	(	PUNCT
cana-5267	215	14	2024	2024	NUM
cana-5267	215	15	)	)	PUNCT
cana-5267	215	16	.	.	PUNCT
cana-5267	216	1	adversarial	adversarial	ADJ
cana-5267	216	2	attacks	attack	NOUN
cana-5267	216	3	and	and	CCONJ
cana-5267	216	4	defenses	defense	NOUN
cana-5267	216	5	in	in	ADP
cana-5267	216	6	nlp	nlp	NOUN
cana-5267	216	7	-	-	PUNCT
cana-5267	216	8	based	base	VERB
cana-5267	216	9	large	large	ADJ
cana-5267	216	10	language	language	NOUN
cana-5267	216	11	models	model	NOUN
cana-5267	216	12	.	.	PUNCT
cana-5267	217	1	ai	ai	PROPN
cana-5267	217	2	&	&	CCONJ
cana-5267	217	3	cybersecurity	cybersecurity	PROPN
cana-5267	217	4	review	review	PROPN
cana-5267	217	5	,	,	PUNCT
cana-5267	217	6	15(1	15(1	NUM
cana-5267	217	7	)	)	PUNCT
cana-5267	217	8	,	,	PUNCT
cana-5267	217	9	55	55	NUM
cana-5267	217	10	-	-	SYM
cana-5267	217	11	78	78	NUM
cana-5267	217	12	.	.	PUNCT
cana-5267	218	1	https://doi.org/10.1007/s10515024-0015	https://doi.org/10.1007/s10515024-0015	PROPN
cana-5267	219	1	[	[	X
cana-5267	219	2	3	3	NUM
cana-5267	219	3	.	.	PUNCT
cana-5267	219	4	]	]	X
cana-5267	220	1	gupta	gupta	PROPN
cana-5267	220	2	,	,	PUNCT
cana-5267	220	3	r.	r.	PROPN
cana-5267	220	4	,	,	PUNCT
cana-5267	220	5	sharma	sharma	PROPN
cana-5267	220	6	,	,	PUNCT
cana-5267	220	7	k.	k.	PROPN
cana-5267	220	8	,	,	PUNCT
cana-5267	220	9	&	&	CCONJ
cana-5267	220	10	singh	singh	PROPN
cana-5267	220	11	,	,	PUNCT
cana-5267	220	12	p.	p.	NOUN
cana-5267	220	13	(	(	PUNCT
cana-5267	220	14	2023	2023	NUM
cana-5267	220	15	)	)	PUNCT
cana-5267	220	16	.	.	PUNCT
cana-5267	221	1	threat	threat	NOUN
cana-5267	221	2	modeling	modeling	NOUN
cana-5267	221	3	for	for	ADP
cana-5267	221	4	ai	ai	ADJ
cana-5267	221	5	-	-	PUNCT
cana-5267	221	6	driven	drive	VERB
cana-5267	221	7	systems	system	NOUN
cana-5267	221	8	:	:	PUNCT
cana-5267	221	9	a	a	DET
cana-5267	221	10	machine	machine	NOUN
cana-5267	221	11	learning	learn	VERB
cana-5267	221	12	perspective	perspective	NOUN
cana-5267	221	13	.	.	PUNCT
cana-5267	222	1	international	international	ADJ
cana-5267	222	2	journal	journal	NOUN
cana-5267	222	3	of	of	ADP
cana-5267	222	4	ai	ai	PROPN
cana-5267	222	5	security	security	NOUN
cana-5267	222	6	,	,	PUNCT
cana-5267	222	7	11(3	11(3	NUM
cana-5267	222	8	)	)	PUNCT
cana-5267	222	9	,	,	PUNCT
cana-5267	222	10	92	92	NUM
cana-5267	222	11	-	-	SYM
cana-5267	222	12	109	109	NUM
cana-5267	222	13	.	.	PUNCT
cana-5267	223	1	https://doi.org/10.1145/3412377	https://doi.org/10.1145/3412377	PUNCT
cana-5267	224	1	[	[	X
cana-5267	224	2	4	4	NUM
cana-5267	224	3	.	.	PUNCT
cana-5267	224	4	]	]	PUNCT
cana-5267	224	5	hall	hall	PROPN
cana-5267	224	6	,	,	PUNCT
cana-5267	224	7	d.	d.	PROPN
cana-5267	224	8	,	,	PUNCT
cana-5267	224	9	&	&	CCONJ
cana-5267	224	10	peterson	peterson	PROPN
cana-5267	224	11	,	,	PUNCT
cana-5267	224	12	r.	r.	PROPN
cana-5267	224	13	(	(	PUNCT
cana-5267	224	14	2022	2022	NUM
cana-5267	224	15	)	)	PUNCT
cana-5267	224	16	.	.	PUNCT
cana-5267	225	1	risk	risk	NOUN
cana-5267	225	2	assessment	assessment	NOUN
cana-5267	225	3	framework	framework	NOUN
cana-5267	225	4	for	for	ADP
cana-5267	225	5	llm	llm	NOUN
cana-5267	225	6	system	system	NOUN
cana-5267	225	7	integration	integration	NOUN
cana-5267	225	8	using	use	VERB
cana-5267	225	9	threat	threat	NOUN
cana-5267	225	10	intelligence	intelligence	NOUN
cana-5267	225	11	.	.	PUNCT
cana-5267	226	1	cybersecurity	cybersecurity	NOUN
cana-5267	226	2	and	and	CCONJ
cana-5267	226	3	ai	ai	PROPN
cana-5267	226	4	journal	journal	PROPN
cana-5267	226	5	,	,	PUNCT
cana-5267	226	6	10(4	10(4	NUM
cana-5267	226	7	)	)	PUNCT
cana-5267	226	8	,	,	PUNCT
cana-5267	226	9	189	189	NUM
cana-5267	226	10	-	-	SYM
cana-5267	226	11	205	205	NUM
cana-5267	226	12	.	.	PUNCT
cana-5267	227	1	https://doi.org/10.1109/csai.2022.003	https://doi.org/10.1109/csai.2022.003	ADJ
cana-5267	227	2	[	[	X
cana-5267	227	3	5	5	NUM
cana-5267	227	4	.	.	PUNCT
cana-5267	227	5	]	]	PUNCT
cana-5267	228	1	kim	kim	PROPN
cana-5267	228	2	,	,	PUNCT
cana-5267	228	3	t.	t.	PROPN
cana-5267	228	4	,	,	PUNCT
cana-5267	228	5	&	&	CCONJ
cana-5267	228	6	zhao	zhao	PROPN
cana-5267	228	7	,	,	PUNCT
cana-5267	228	8	y.	y.	PROPN
cana-5267	228	9	(	(	PUNCT
cana-5267	228	10	2023	2023	NUM
cana-5267	228	11	)	)	PUNCT
cana-5267	228	12	.	.	PUNCT
cana-5267	229	1	mitigating	mitigate	VERB
cana-5267	229	2	security	security	NOUN
cana-5267	229	3	risks	risk	NOUN
cana-5267	229	4	in	in	ADP
cana-5267	229	5	large	large	ADJ
cana-5267	229	6	language	language	NOUN
cana-5267	229	7	models	model	NOUN
cana-5267	229	8	through	through	ADP
cana-5267	229	9	automated	automate	VERB
cana-5267	229	10	threat	threat	NOUN
cana-5267	229	11	detection	detection	NOUN
cana-5267	229	12	.	.	PUNCT
cana-5267	230	1	computational	computational	ADJ
cana-5267	230	2	security	security	NOUN
cana-5267	230	3	review	review	NOUN
cana-5267	230	4	,	,	PUNCT
cana-5267	230	5	8(2	8(2	NUM
cana-5267	230	6	)	)	PUNCT
cana-5267	230	7	,	,	PUNCT
cana-5267	230	8	75	75	NUM
cana-5267	230	9	-	-	SYM
cana-5267	230	10	98	98	NUM
cana-5267	230	11	.	.	PUNCT
cana-5267	231	1	https://doi.org/10.1186/s40537-023-00234	https://doi.org/10.1186/s40537-023-00234	NOUN
cana-5267	232	1	[	[	X
cana-5267	232	2	6	6	NUM
cana-5267	232	3	.	.	PUNCT
cana-5267	232	4	]	]	X
cana-5267	232	5	lee	lee	PROPN
cana-5267	232	6	,	,	PUNCT
cana-5267	232	7	c.	c.	PROPN
cana-5267	232	8	,	,	PUNCT
cana-5267	232	9	&	&	CCONJ
cana-5267	232	10	park	park	PROPN
cana-5267	232	11	,	,	PUNCT
cana-5267	232	12	j.	j.	PROPN
cana-5267	232	13	(	(	PUNCT
cana-5267	232	14	2024	2024	NUM
cana-5267	232	15	)	)	PUNCT
cana-5267	232	16	.	.	PUNCT
cana-5267	233	1	prompt	prompt	ADJ
cana-5267	233	2	injection	injection	NOUN
cana-5267	233	3	attacks	attack	NOUN
cana-5267	233	4	:	:	PUNCT
cana-5267	233	5	an	an	DET
cana-5267	233	6	emerging	emerge	VERB
cana-5267	233	7	threat	threat	NOUN
cana-5267	233	8	in	in	ADP
cana-5267	233	9	nlp	nlp	NOUN
cana-5267	233	10	systems	system	NOUN
cana-5267	233	11	.	.	PUNCT
cana-5267	234	1	ai	ai	PROPN
cana-5267	234	2	&	&	CCONJ
cana-5267	234	3	data	data	PROPN
cana-5267	234	4	privacy	privacy	PROPN
cana-5267	234	5	journal	journal	PROPN
cana-5267	234	6	,	,	PUNCT
cana-5267	234	7	12(1	12(1	NUM
cana-5267	234	8	)	)	PUNCT
cana-5267	234	9	,	,	PUNCT
cana-5267	234	10	67	67	NUM
cana-5267	234	11	-	-	SYM
cana-5267	234	12	83	83	NUM
cana-5267	234	13	.	.	PUNCT
cana-5267	235	1	https://doi.org/10.1016/j.priv.2024.005	https://doi.org/10.1016/j.priv.2024.005	PROPN
cana-5267	236	1	[	[	X
cana-5267	236	2	7	7	NUM
cana-5267	236	3	.	.	PUNCT
cana-5267	236	4	]	]	X
cana-5267	237	1	li	li	PROPN
cana-5267	237	2	,	,	PUNCT
cana-5267	237	3	h.	h.	PROPN
cana-5267	237	4	,	,	PUNCT
cana-5267	237	5	sun	sun	PROPN
cana-5267	237	6	,	,	PUNCT
cana-5267	237	7	g.	g.	PROPN
cana-5267	237	8	,	,	PUNCT
cana-5267	237	9	&	&	CCONJ
cana-5267	237	10	zhao	zhao	NOUN
cana-5267	237	11	,	,	PUNCT
cana-5267	237	12	m.	m.	NOUN
cana-5267	237	13	(	(	PUNCT
cana-5267	237	14	2023	2023	NUM
cana-5267	237	15	)	)	PUNCT
cana-5267	237	16	.	.	PUNCT
cana-5267	238	1	machine	machine	NOUN
cana-5267	238	2	learning	learning	NOUN
cana-5267	238	3	-	-	PUNCT
cana-5267	238	4	based	base	VERB
cana-5267	238	5	anomaly	anomaly	NOUN
cana-5267	238	6	detection	detection	NOUN
cana-5267	238	7	in	in	ADP
cana-5267	238	8	largescale	largescale	ADJ
cana-5267	238	9	nlp	nlp	NOUN
cana-5267	238	10	deployments	deployment	NOUN
cana-5267	238	11	.	.	PUNCT
cana-5267	239	1	journal	journal	NOUN
cana-5267	239	2	of	of	ADP
cana-5267	239	3	ai	ai	PROPN
cana-5267	239	4	risk	risk	NOUN
cana-5267	239	5	analysis	analysis	NOUN
cana-5267	239	6	,	,	PUNCT
cana-5267	239	7	14(2	14(2	NUM
cana-5267	239	8	)	)	PUNCT
cana-5267	239	9	,	,	PUNCT
cana-5267	239	10	130	130	NUM
cana-5267	239	11	-	-	SYM
cana-5267	239	12	152	152	NUM
cana-5267	239	13	.	.	PUNCT
cana-5267	240	1	https://doi.org/10.1109/aira.2023.004	https://doi.org/10.1109/aira.2023.004	NOUN
cana-5267	240	2	[	[	X
cana-5267	240	3	8	8	NUM
cana-5267	240	4	.	.	PUNCT
cana-5267	240	5	]	]	X
cana-5267	241	1	miller	miller	PROPN
cana-5267	241	2	,	,	PUNCT
cana-5267	241	3	j.	j.	PROPN
cana-5267	241	4	,	,	PUNCT
cana-5267	241	5	&	&	CCONJ
cana-5267	241	6	davis	davis	PROPN
cana-5267	241	7	,	,	PUNCT
cana-5267	241	8	l.	l.	PROPN
cana-5267	241	9	(	(	PUNCT
cana-5267	241	10	2022	2022	NUM
cana-5267	241	11	)	)	PUNCT
cana-5267	241	12	.	.	PUNCT
cana-5267	242	1	enhancing	enhance	VERB
cana-5267	242	2	llm	llm	PROPN
cana-5267	242	3	security	security	NOUN
cana-5267	242	4	through	through	ADP
cana-5267	242	5	automated	automate	VERB
cana-5267	242	6	adversarial	adversarial	ADJ
cana-5267	242	7	testing	testing	NOUN
cana-5267	242	8	.	.	PUNCT
cana-5267	243	1	journal	journal	NOUN
cana-5267	243	2	of	of	ADP
cana-5267	243	3	machine	machine	NOUN
cana-5267	243	4	learning	learn	VERB
cana-5267	243	5	security	security	NOUN
cana-5267	243	6	,	,	PUNCT
cana-5267	243	7	7(3	7(3	NUM
cana-5267	243	8	)	)	PUNCT
cana-5267	243	9	,	,	PUNCT
cana-5267	243	10	155	155	NUM
cana-5267	243	11	-	-	SYM
cana-5267	243	12	172	172	NUM
cana-5267	243	13	.	.	PUNCT
cana-5267	244	1	communications	communication	NOUN
cana-5267	244	2	on	on	ADP
cana-5267	244	3	applied	apply	VERB
cana-5267	244	4	nonlinear	nonlinear	ADJ
cana-5267	244	5	analysis	analysis	NOUN
cana-5267	244	6	issn	issn	NOUN
cana-5267	244	7	:	:	PUNCT
cana-5267	244	8	1074	1074	NUM
cana-5267	244	9	-	-	PUNCT
cana-5267	244	10	133x	133x	NUM
cana-5267	244	11	vol	vol	VERB
cana-5267	244	12	32	32	NUM
cana-5267	244	13	no	no	NOUN
cana-5267	244	14	.	.	PUNCT
cana-5267	245	1	10s	10	NOUN
cana-5267	245	2	(	(	PUNCT
cana-5267	245	3	2025	2025	NUM
cana-5267	245	4	)	)	PUNCT
cana-5267	245	5	1670	1670	NUM
cana-5267	245	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5267	246	1	https://doi.org/10.1016/j.mlsec.2022.007	https://doi.org/10.1016/j.mlsec.2022.007	NOUN
cana-5267	247	1	[	[	X
cana-5267	247	2	9	9	NUM
cana-5267	247	3	.	.	PUNCT
cana-5267	247	4	]	]	PUNCT
cana-5267	248	1	nguyen	nguyen	NOUN
cana-5267	248	2	,	,	PUNCT
cana-5267	248	3	t.	t.	PROPN
cana-5267	248	4	,	,	PUNCT
cana-5267	248	5	&	&	CCONJ
cana-5267	248	6	bui	bui	PROPN
cana-5267	248	7	,	,	PUNCT
cana-5267	248	8	h.	h.	PROPN
cana-5267	248	9	(	(	PUNCT
cana-5267	248	10	2024	2024	NUM
cana-5267	248	11	)	)	PUNCT
cana-5267	248	12	.	.	PUNCT
cana-5267	249	1	cybersecurity	cybersecurity	NOUN
cana-5267	249	2	challenges	challenge	NOUN
cana-5267	249	3	in	in	ADP
cana-5267	249	4	nlp	nlp	NOUN
cana-5267	249	5	-	-	PUNCT
cana-5267	249	6	driven	drive	VERB
cana-5267	249	7	ai	ai	NOUN
cana-5267	249	8	models	model	NOUN
cana-5267	249	9	:	:	PUNCT
cana-5267	249	10	a	a	DET
cana-5267	249	11	review	review	NOUN
cana-5267	249	12	.	.	PUNCT
cana-5267	250	1	journal	journal	NOUN
cana-5267	250	2	of	of	ADP
cana-5267	250	3	computational	computational	ADJ
cana-5267	250	4	security	security	NOUN
cana-5267	250	5	&	&	CCONJ
cana-5267	250	6	intelligence	intelligence	NOUN
cana-5267	250	7	,	,	PUNCT
cana-5267	250	8	16(1	16(1	NUM
cana-5267	250	9	)	)	PUNCT
cana-5267	250	10	,	,	PUNCT
cana-5267	250	11	98	98	NUM
cana-5267	250	12	-	-	SYM
cana-5267	250	13	121	121	NUM
cana-5267	250	14	.	.	PUNCT
cana-5267	251	1	https://doi.org/10.1109/jcsi.2024.006	https://doi.org/10.1109/jcsi.2024.006	PROPN
cana-5267	251	2	[	[	X
cana-5267	251	3	10	10	NUM
cana-5267	251	4	.	.	PUNCT
cana-5267	251	5	]	]	PUNCT
cana-5267	252	1	patel	patel	PROPN
cana-5267	252	2	,	,	PUNCT
cana-5267	252	3	m.	m.	NOUN
cana-5267	252	4	,	,	PUNCT
cana-5267	252	5	&	&	CCONJ
cana-5267	252	6	shah	shah	PROPN
cana-5267	252	7	,	,	PUNCT
cana-5267	252	8	d.	d.	PROPN
cana-5267	252	9	(	(	PUNCT
cana-5267	252	10	2023	2023	NUM
cana-5267	252	11	)	)	PUNCT
cana-5267	252	12	.	.	PUNCT
cana-5267	253	1	automated	automate	VERB
cana-5267	253	2	threat	threat	NOUN
cana-5267	253	3	intelligence	intelligence	NOUN
cana-5267	253	4	for	for	ADP
cana-5267	253	5	llm	llm	PROPN
cana-5267	253	6	systems	system	NOUN
cana-5267	253	7	:	:	PUNCT
cana-5267	253	8	a	a	DET
cana-5267	253	9	hybrid	hybrid	ADJ
cana-5267	253	10	machine	machine	NOUN
cana-5267	253	11	learning	learning	NOUN
cana-5267	253	12	approach	approach	NOUN
cana-5267	253	13	.	.	PUNCT
cana-5267	254	1	international	international	ADJ
cana-5267	254	2	journal	journal	PROPN
cana-5267	254	3	of	of	ADP
cana-5267	254	4	ai	ai	PROPN
cana-5267	254	5	&	&	CCONJ
cana-5267	254	6	cyber	cyber	PROPN
cana-5267	254	7	threats	threat	NOUN
cana-5267	254	8	,	,	PUNCT
cana-5267	254	9	9(2	9(2	NUM
cana-5267	254	10	)	)	PUNCT
cana-5267	254	11	,	,	PUNCT
cana-5267	254	12	85	85	NUM
cana-5267	254	13	-	-	SYM
cana-5267	254	14	101	101	NUM
cana-5267	254	15	.	.	PUNCT
cana-5267	255	1	https://doi.org/10.1109/ijaict.2023.009	https://doi.org/10.1109/ijaict.2023.009	PROPN
cana-5267	255	2	[	[	X
cana-5267	255	3	11	11	NUM
cana-5267	255	4	.	.	PUNCT
cana-5267	255	5	]	]	PUNCT
cana-5267	256	1	qureshi	qureshi	PROPN
cana-5267	256	2	,	,	PUNCT
cana-5267	256	3	f.	f.	PROPN
cana-5267	256	4	,	,	PUNCT
cana-5267	256	5	&	&	CCONJ
cana-5267	256	6	khan	khan	PROPN
cana-5267	256	7	,	,	PUNCT
cana-5267	256	8	a.	a.	NOUN
cana-5267	256	9	(	(	PUNCT
cana-5267	256	10	2024	2024	NUM
cana-5267	256	11	)	)	PUNCT
cana-5267	256	12	.	.	PUNCT
cana-5267	257	1	real	real	ADJ
cana-5267	257	2	-	-	PUNCT
cana-5267	257	3	time	time	NOUN
cana-5267	257	4	security	security	NOUN
cana-5267	257	5	monitoring	monitoring	NOUN
cana-5267	257	6	in	in	ADP
cana-5267	257	7	ai	ai	ADJ
cana-5267	257	8	-	-	PUNCT
cana-5267	257	9	powered	power	VERB
cana-5267	257	10	language	language	NOUN
cana-5267	257	11	models	model	NOUN
cana-5267	257	12	.	.	PUNCT
cana-5267	258	1	computational	computational	ADJ
cana-5267	258	2	intelligence	intelligence	PROPN
cana-5267	258	3	&	&	CCONJ
cana-5267	258	4	security	security	PROPN
cana-5267	258	5	journal	journal	NOUN
cana-5267	258	6	,	,	PUNCT
cana-5267	258	7	13(1	13(1	NUM
cana-5267	258	8	)	)	PUNCT
cana-5267	258	9	,	,	PUNCT
cana-5267	258	10	112	112	NUM
cana-5267	258	11	-	-	SYM
cana-5267	258	12	136	136	NUM
cana-5267	258	13	.	.	PUNCT
cana-5267	259	1	https://doi.org/10.1007/s11042-024-0123	https://doi.org/10.1007/s11042-024-0123	VERB
cana-5267	259	2	[	[	X
cana-5267	259	3	12	12	NUM
cana-5267	259	4	.	.	PUNCT
cana-5267	259	5	]	]	X
cana-5267	260	1	raj	raj	PROPN
cana-5267	260	2	,	,	PUNCT
cana-5267	260	3	p.	p.	PROPN
cana-5267	260	4	,	,	PUNCT
cana-5267	260	5	&	&	CCONJ
cana-5267	260	6	kumar	kumar	PROPN
cana-5267	260	7	,	,	PUNCT
cana-5267	260	8	s.	s.	PROPN
cana-5267	260	9	(	(	PUNCT
cana-5267	260	10	2023	2023	NUM
cana-5267	260	11	)	)	PUNCT
cana-5267	260	12	.	.	PUNCT
cana-5267	261	1	defensive	defensive	ADJ
cana-5267	261	2	strategies	strategy	NOUN
cana-5267	261	3	against	against	ADP
cana-5267	261	4	model	model	NOUN
cana-5267	261	5	inversion	inversion	NOUN
cana-5267	261	6	attacks	attack	NOUN
cana-5267	261	7	in	in	ADP
cana-5267	261	8	llms	llm	NOUN
cana-5267	261	9	.	.	PUNCT
cana-5267	262	1	ai	ai	VERB
cana-5267	262	2	security	security	NOUN
cana-5267	262	3	research	research	NOUN
cana-5267	262	4	journal	journal	NOUN
cana-5267	262	5	,	,	PUNCT
cana-5267	262	6	10(3	10(3	NUM
cana-5267	262	7	)	)	PUNCT
cana-5267	262	8	,	,	PUNCT
cana-5267	262	9	204	204	NUM
cana-5267	262	10	-	-	SYM
cana-5267	262	11	219	219	NUM
cana-5267	262	12	.	.	PUNCT
cana-5267	263	1	https://doi.org/10.1016/j.aisr.2023.008	https://doi.org/10.1016/j.aisr.2023.008	NOUN
cana-5267	264	1	[	[	X
cana-5267	264	2	13	13	NUM
cana-5267	264	3	.	.	PUNCT
cana-5267	264	4	]	]	PUNCT
cana-5267	265	1	smith	smith	PROPN
cana-5267	265	2	,	,	PUNCT
cana-5267	265	3	e.	e.	PROPN
cana-5267	265	4	,	,	PUNCT
cana-5267	265	5	&	&	CCONJ
cana-5267	265	6	taylor	taylor	PROPN
cana-5267	265	7	,	,	PUNCT
cana-5267	265	8	b.	b.	PROPN
cana-5267	265	9	(	(	PUNCT
cana-5267	265	10	2022	2022	NUM
cana-5267	265	11	)	)	PUNCT
cana-5267	265	12	.	.	PUNCT
cana-5267	266	1	integrating	integrate	VERB
cana-5267	266	2	automated	automate	VERB
cana-5267	266	3	threat	threat	NOUN
cana-5267	266	4	detection	detection	NOUN
cana-5267	266	5	in	in	ADP
cana-5267	266	6	natural	natural	ADJ
cana-5267	266	7	language	language	NOUN
cana-5267	266	8	processing	processing	NOUN
cana-5267	266	9	pipelines	pipeline	NOUN
cana-5267	266	10	.	.	PUNCT
cana-5267	267	1	ai	ai	PROPN
cana-5267	267	2	&	&	CCONJ
cana-5267	267	3	cybersecurity	cybersecurity	NOUN
cana-5267	267	4	transactions	transaction	NOUN
cana-5267	267	5	,	,	PUNCT
cana-5267	267	6	11(2	11(2	NUM
cana-5267	267	7	)	)	PUNCT
cana-5267	267	8	,	,	PUNCT
cana-5267	267	9	75	75	NUM
cana-5267	267	10	-	-	SYM
cana-5267	267	11	94	94	NUM
cana-5267	267	12	.	.	PUNCT
cana-5267	268	1	https://doi.org/10.1016/j.aics.2022.005	https://doi.org/10.1016/j.aics.2022.005	PROPN
cana-5267	269	1	[	[	X
cana-5267	269	2	14	14	NUM
cana-5267	269	3	.	.	PUNCT
cana-5267	269	4	]	]	PUNCT
cana-5267	270	1	wang	wang	PROPN
cana-5267	270	2	,	,	PUNCT
cana-5267	270	3	x.	x.	PROPN
cana-5267	270	4	,	,	PUNCT
cana-5267	270	5	&	&	CCONJ
cana-5267	270	6	xu	xu	PROPN
cana-5267	270	7	,	,	PUNCT
cana-5267	270	8	l.	l.	PROPN
cana-5267	270	9	(	(	PUNCT
cana-5267	270	10	2023	2023	NUM
cana-5267	270	11	)	)	PUNCT
cana-5267	270	12	.	.	PUNCT
cana-5267	271	1	deep	deep	ADJ
cana-5267	271	2	learning	learning	NOUN
cana-5267	271	3	approaches	approach	NOUN
cana-5267	271	4	to	to	PART
cana-5267	271	5	threat	threat	NOUN
cana-5267	271	6	modeling	modeling	NOUN
cana-5267	271	7	in	in	ADP
cana-5267	271	8	large	large	ADJ
cana-5267	271	9	-	-	PUNCT
cana-5267	271	10	scale	scale	NOUN
cana-5267	271	11	ai	ai	NOUN
cana-5267	271	12	systems	system	NOUN
cana-5267	271	13	.	.	PUNCT
cana-5267	272	1	journal	journal	PROPN
cana-5267	272	2	of	of	ADP
cana-5267	272	3	ai	ai	PROPN
cana-5267	272	4	-	-	PUNCT
cana-5267	272	5	driven	drive	VERB
cana-5267	272	6	security	security	NOUN
cana-5267	272	7	solutions	solution	NOUN
cana-5267	272	8	,	,	PUNCT
cana-5267	272	9	12(4	12(4	NUM
cana-5267	272	10	)	)	PUNCT
cana-5267	272	11	,	,	PUNCT
cana-5267	272	12	147	147	NUM
cana-5267	272	13	-	-	SYM
cana-5267	272	14	167	167	NUM
cana-5267	272	15	.	.	PUNCT
cana-5267	273	1	https://doi.org/10.1109/jaiss.2023.015	https://doi.org/10.1109/jaiss.2023.015	PROPN
cana-5267	273	2	[	[	X
cana-5267	273	3	15	15	NUM
cana-5267	273	4	.	.	PUNCT
cana-5267	273	5	]	]	X
cana-5267	274	1	zhou	zhou	PROPN
cana-5267	274	2	,	,	PUNCT
cana-5267	274	3	p.	p.	PROPN
cana-5267	274	4	,	,	PUNCT
cana-5267	274	5	&	&	CCONJ
cana-5267	274	6	lin	lin	PROPN
cana-5267	274	7	,	,	PUNCT
cana-5267	274	8	k.	k.	PROPN
cana-5267	274	9	(	(	PUNCT
cana-5267	274	10	2024	2024	NUM
cana-5267	274	11	)	)	PUNCT
cana-5267	274	12	.	.	PUNCT
cana-5267	275	1	exploring	explore	VERB
cana-5267	275	2	the	the	DET
cana-5267	275	3	use	use	NOUN
cana-5267	275	4	of	of	ADP
cana-5267	275	5	nlp	nlp	NOUN
cana-5267	275	6	for	for	ADP
cana-5267	275	7	threat	threat	NOUN
cana-5267	275	8	intelligence	intelligence	NOUN
cana-5267	275	9	in	in	ADP
cana-5267	275	10	ai	ai	PRON
cana-5267	275	11	security	security	NOUN
cana-5267	275	12	.	.	PUNCT
cana-5267	276	1	machine	machine	NOUN
cana-5267	276	2	learning	learning	PROPN
cana-5267	276	3	&	&	CCONJ
cana-5267	276	4	security	security	PROPN
cana-5267	276	5	review	review	PROPN
cana-5267	276	6	,	,	PUNCT
cana-5267	276	7	15(1	15(1	NUM
cana-5267	276	8	)	)	PUNCT
cana-5267	276	9	,	,	PUNCT
cana-5267	276	10	178	178	NUM
cana-5267	276	11	-	-	SYM
cana-5267	276	12	196	196	NUM
cana-5267	276	13	.	.	PUNCT
cana-5267	277	1	https://doi.org/10.1016/j.mlsr.2024.009	https://doi.org/10.1016/j.mlsr.2024.009	PROPN