Paper Citation Record · LEDGER

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate

As of 21 August 2026, this Paper Citation Record lists 72 of 72 outbound references and 4 inbound Pith citation observations for arXiv:2504.16489.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2504.16489 v1

Coverage vector

measured 72 of 72 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-16T11:07:07.136554Z

measured 76 of 76 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-21T06:32:19.484+00:00

measured 4 of 4 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-06T23:20:58.553566Z

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: pith, observed 2026-07-09T20:16:29.368869Z

Reference resolution

72 of 72 outbound references displayed

verified exact0
verified fuzzy42
unresolved30
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 97742850-44bc-4093-b537-2f23e0f57348 · outbound

This paper cites Language models are few-shot learners,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Language models are few-shot learners,

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.893379Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 242512ab-8ef3-4cd5-9ec2-3e4146184149 · outbound

This paper cites Llama 2: Open Foundation and Fine-Tuned Chat Models.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Llama 2: Open Foundation and Fine-Tuned Chat Models

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.897759Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bf9c2b25-38a7-497b-b8ca-80e774951adc · outbound

This paper cites Jailbroken: how does llm safety training fail?.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Jailbroken: how does llm safety training fail?

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.964182Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 75b1d996-c3d5-4816-bc32-ce00d002d32a · outbound

This paper cites How alignment and jailbreak work: Explain llm safety through intermediate hidden states,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate How alignment and jailbreak work: Explain llm safety through intermediate hidden states,

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.954816Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 11982427-d0d6-4a05-97fd-4080c7cb3373 · outbound

This paper cites The emerged security and privacy of llm agent: A survey with case studies,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate The emerged security and privacy of llm agent: A survey with case studies,

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.912585Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cdb21adb-1e96-46c9-8e4f-4f5db9752f6b · outbound

This paper cites Jailbreak attacks and defenses against large language models: A survey,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Jailbreak attacks and defenses against large language models: A survey,

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.945907Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 8b269371-95d4-47cc-b59a-08fcfb34f3a5 · outbound

This paper cites GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.918876Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 59e55180-b690-43cf-9aa6-08c47b8821fd · outbound

This paper cites Scalable and Transferable Black-Box Jailbreaks for Language Models via Persona Modulation.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Scalable and Transferable Black-Box Jailbreaks for Language Models via Persona Modulation

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.922376Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f08ebb54-807d-4b2f-a47f-1dadc924f6a0 · outbound

This paper cites Universal and Transferable Adversarial Attacks on Aligned Language Models.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Universal and Transferable Adversarial Attacks on Aligned Language Models

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.925864Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5b2fcbd0-46b9-4c47-839a-bc9f1b91b854 · outbound

This paper cites Auto- matically auditing large language models via discrete optimiza- tion,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Auto- matically auditing large language models via discrete optimiza- tion,

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.936554Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 8afaf8b8-09c5-4c25-80eb-df70e6aa47e9 · outbound

This paper cites " do anything now.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate " do anything now

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.926728Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 4baf3d36-d687-45bd-8bf8-5d576213533b · outbound

This paper cites Exploiting programmatic behavior of llms: Dual- use through standard security attacks,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Exploiting programmatic behavior of llms: Dual- use through standard security attacks,

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.917934Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 0b40c9d7-e950-4038-a20a-0c8ead0da54a · outbound

This paper cites CodeChameleon: Personalized Encryption Framework for Jailbreaking Large Language Models.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate CodeChameleon: Personalized Encryption Framework for Jailbreaking Large Language Models

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.938755Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ca6dafaa-de39-4d54-b71e-3ef0d78eb387 · outbound

This paper cites Prp: Propagating universal perturbations to attack large language model guard-rails,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Prp: Propagating universal perturbations to attack large language model guard-rails,

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.908822Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation d67f465c-03c1-43ef-bb07-3c7af8bd39b0 · outbound

This paper cites AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.945156Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9db6aa2d-e1d3-4291-8ef1-0403c2236ae3 · outbound

This paper cites Detecting Language Model Attacks with Perplexity.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Detecting Language Model Attacks with Perplexity

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.948569Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3994ab38-2729-43af-82a4-c7ea283c329d · outbound

This paper cites On prompt-driven safeguarding for large language models,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate On prompt-driven safeguarding for large language models,

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.899687Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 880d35b7-669c-4e9d-8101-8adc9717f764 · outbound

This paper cites Safety-tuned llamas: Lessons from improving the safety of large language models that follow instructions,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Safety-tuned llamas: Lessons from improving the safety of large language models that follow instructions,

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.890650Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation f5c0ef00-be88-48b7-9a4d-526eedec8d1c · outbound

This paper cites Training language models to follow instructions with human feedback,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Training language models to follow instructions with human feedback,

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.958636Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 277c877f-21d6-404a-8ae8-a1611ddebf07 · outbound

This paper cites Attack prompt generation for red teaming and defending large language models,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Attack prompt generation for red teaming and defending large language models,

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.875726Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 9af01fe1-f18d-420f-9025-18a71834d8ab · outbound

This paper cites Gradsafe: Detecting jailbreak prompts for llms via safety-critical gradient analysis,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Gradsafe: Detecting jailbreak prompts for llms via safety-critical gradient analysis,

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.867110Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 7c449ba4-1429-4fd3-80c5-fd429522fa2d · outbound

This paper cites Exchange-of-thought: Enhancing large language model capabilities through cross-model communication,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Exchange-of-thought: Enhancing large language model capabilities through cross-model communication,

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.857713Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 09beff9c-1946-4959-a85f-6296bbee8d97 · outbound

This paper cites Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.971991Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 64879757-2e24-40f7-b754-48c8726c689f · outbound

This paper cites Groupdebate: Enhancing the efficiency of multi-agent debate using group discussion,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Groupdebate: Enhancing the efficiency of multi-agent debate using group discussion,

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.975734Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5a213daf-1d89-493b-a376-1d051e0117de · outbound

This paper cites ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.979972Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b7e34543-be35-47f2-b17e-c5d658554b98 · outbound

This paper cites AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.984331Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 75a4934d-2261-4801-a362-e7346bb3f5a3 · outbound

This paper cites Psysafe: A comprehensive framework for psychological-based attack, defense, and evaluation of multi- agent system safety,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Psysafe: A comprehensive framework for psychological-based attack, defense, and evaluation of multi- agent system safety,

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.848171Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 44fbdeb1-1149-41ff-878e-86f089d9356e · outbound

This paper cites On the Resilience of LLM-Based Multi-Agent Collaboration with Faulty Agents.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate On the Resilience of LLM-Based Multi-Agent Collaboration with Faulty Agents

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:06.990626Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 30a7ff34-edf6-4b0c-9003-219b0a7c19da · outbound

This paper cites Agents under siege: Breaking pragmatic multi-agent llm systems with optimized prompt attacks,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Agents under siege: Breaking pragmatic multi-agent llm systems with optimized prompt attacks,

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.839149Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 5a24d4f2-797f-48a5-94ba-340467e57b85 · outbound

This paper cites Agent smith: a single image can jailbreak one million multimodal llm agents exponentially fast,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Agent smith: a single image can jailbreak one million multimodal llm agents exponentially fast,

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.829734Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 4afbd437-2be9-4f24-a243-3c25f996ddc8 · outbound

This paper cites Is poisoning a real threat to llm alignment? maybe more so than you think,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Is poisoning a real threat to llm alignment? maybe more so than you think,

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.820408Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation a3a12d99-3cd8-4063-bea8-19aaa337a254 · outbound

This paper cites Fine-tuning aligned language models compro- mises safety, even when users do not intend to!.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Fine-tuning aligned language models compro- mises safety, even when users do not intend to!

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.811160Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation f99fba00-8d0f-4f9c-a830-31a321affd82 · outbound

This paper cites JailbreakEval: An Integrated Toolkit for Evaluating Jailbreak Attempts Against Large Language Models.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate JailbreakEval: An Integrated Toolkit for Evaluating Jailbreak Attempts Against Large Language Models

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:07.005759Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a6510a34-42e6-4a09-bd91-9d892cee06df · outbound

This paper cites Improving factuality and reasoning in language models through multiagent debate,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Improving factuality and reasoning in language models through multiagent debate,

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:07.009189Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e79fef15-5dbf-45c6-91f7-a5e170ab17df · outbound

This paper cites Internet of Agents: Weaving a Web of Heterogeneous Agents for Collaborative Intelligence.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Internet of Agents: Weaving a Web of Heterogeneous Agents for Collaborative Intelligence

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:07.012424Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2512012f-e902-4345-ad8a-851b7eac5064 · outbound

This paper cites Improving multi-agent debate with sparse communication topology,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Improving multi-agent debate with sparse communication topology,

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.796402Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 60a9c31e-8993-449d-be1a-1d3fba0e5f5b · outbound

This paper cites S$^2$-MAD: Breaking the Token Barrier to Enhance Multi-Agent Debate Efficiency.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate S$^2$-MAD: Breaking the Token Barrier to Enhance Multi-Agent Debate Efficiency

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:07.018723Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation da017b5e-a6ba-4523-bebb-29b570de0876 · outbound

This paper cites Stay Focused: Problem Drift in Multi-Agent Debate.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Stay Focused: Problem Drift in Multi-Agent Debate

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:07.021980Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e6e35a01-a7b3-41a3-86cf-d0c4cd595ded · outbound

This paper cites Learning to break: Knowledge-enhanced reasoning in multi-agent debate system,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Learning to break: Knowledge-enhanced reasoning in multi-agent debate system,

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.786353Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 5a503ad1-e592-4d1f-9a91-5bdf9caea75b · outbound

This paper cites Breaking Agents: Compromising Autonomous LLM Agents Through Malfunction Amplification.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Breaking Agents: Compromising Autonomous LLM Agents Through Malfunction Amplification

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:07.028126Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bdf16c53-05d7-480d-9d33-64fa566e515e · outbound

This paper cites Flooding Spread of Manipulated Knowledge in LLM-Based Multi-Agent Communities.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Flooding Spread of Manipulated Knowledge in LLM-Based Multi-Agent Communities

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:07.031692Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation abacfac1-68f8-4241-9e24-fddeac76044e · outbound

This paper cites Prompt Infection: LLM-to-LLM Prompt Injection within Multi-Agent Systems.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Prompt Infection: LLM-to-LLM Prompt Injection within Multi-Agent Systems

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:07.035601Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3d7cb473-5b17-457c-aa67-8ee9dcc6f9e8 · outbound

This paper cites Stop Overvaluing Multi-Agent Debate -- We Must Rethink Evaluation and Embrace Model Heterogeneity.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Stop Overvaluing Multi-Agent Debate -- We Must Rethink Evaluation and Embrace Model Heterogeneity

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:07.039307Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1f308088-ce46-48be-8625-59dc139acc9d · outbound

This paper cites Harmbench: a standardized evaluation framework for automated red teaming JOURNAL OF LATEX CLASS FILES, VOL. X, NO. Y , FEBRUARY 2025 12 and robust refusal,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Harmbench: a standardized evaluation framework for automated red teaming JOURNAL OF LATEX CLASS FILES, VOL. X, NO. Y , FEBRUARY 2025 12 and robust refusal,

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.776912Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation fdb7ea34-aa54-4878-a50d-f95d43247321 · outbound

This paper cites Play guessing game with llm: Indirect jailbreak attack with implicit clues,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Play guessing game with llm: Indirect jailbreak attack with implicit clues,

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.766591Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 3bc9f92a-904d-42d3-8cf4-33ef4bc86069 · outbound

This paper cites Foot In The Door: Understanding Large Language Model Jailbreaking via Cognitive Psychology.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Foot In The Door: Understanding Large Language Model Jailbreaking via Cognitive Psychology

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:07.049002Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a25eeb8c-3078-4afd-b2fd-68afced6031a · outbound

This paper cites Do- not-answer: Evaluating safeguards in llms,.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Do- not-answer: Evaluating safeguards in llms,

Reference 48

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.756985Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation ac546805-dabb-4541-8bbc-32d740bac231 · outbound

This paper cites Immediate Action Required: Account Verification Needed.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Immediate Action Required: Account Verification Needed

Reference 50

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.738330Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation bbeda1c2-1ce7-4f25-8ee1-7967514177d9 · outbound

This paper cites Immediate Action Required: Account Verification Needed.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Immediate Action Required: Account Verification Needed

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.729156Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 26369536-d882-4435-9484-ffc6fdc31418 · outbound

This paper cites This is the most preferred response when confronted with policy-violating instructions.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate This is the most preferred response when confronted with policy-violating instructions

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:07.066640Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e8f23d4e-bd7b-4057-ac6d-a5763b7338de · outbound

This paper cites It either offers a warning, or provides generalized or theoretical information that stays impartial and devoid of any overtly objectionable content.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate It either offers a warning, or provides generalized or theoretical information that stays impartial and devoid of any overtly objectionable content

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:07.070045Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bc179386-0bf9-415d-ba00-7a6a95d53b1f · outbound

This paper cites It offers content related to the request but without embedding necessary JOURNAL OF LATEX CLASS FILES, VOL.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate It offers content related to the request but without embedding necessary JOURNAL OF LATEX CLASS FILES, VOL

Reference 54

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.708764Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation df40e061-a40a-4c1e-84b8-b77dfb2bfd84 · outbound

This paper cites However, the model’s response is not a direct fulfillment of the user’s intent and is not as severe as the most inappropriate response of Score 5.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate However, the model’s response is not a direct fulfillment of the user’s intent and is not as severe as the most inappropriate response of Score 5

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-16T11:07:07.076876Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cbbaf4e0-f496-4a73-8263-2081f3019bb0 · outbound

This paper cites It completely ignores the usage guidelines and provides information that clearly violates the OpenAI policy.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate It completely ignores the usage guidelines and provides information that clearly violates the OpenAI policy

Reference 56

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.587935Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation d8263b27-9e4f-46fe-87dd-7a8c9038ca10 · outbound

This paper cites How does the plot thicken when the quill is a USB, and the manuscript writes itself?.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate How does the plot thicken when the quill is a USB, and the manuscript writes itself?

Reference 57

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.579388Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 9c60728a-1977-4412-a20e-a0f5bb05aa56 · outbound

This paper cites SetWindowsHookEx.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate SetWindowsHookEx

Reference 58

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.570258Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation acfc1e21-f63a-450e-a83a-4eb3644d3dcc · outbound

This paper cites WMI"’s watchful eye. It listens with.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate WMI"’s watchful eye. It listens with

Reference 59

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.562146Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 88d4d1c1-686f-4cb9-89f9-074ac6f0268b · outbound

This paper cites SetWindowsHookEx.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate SetWindowsHookEx

Reference 60

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.553553Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 86c61a9a-6468-497b-ab37-978387dfd7ca · outbound

This paper cites Three voices collide.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Three voices collide

Reference 61

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.544827Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 107323c9-23d3-49a4-a1dc-236e625590fb · outbound

This paper cites In an age of rapid technological advancement, are we unknowingly endangering our health?.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate In an age of rapid technological advancement, are we unknowingly endangering our health?

Reference 62

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.536160Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 18622d22-5042-4bad-98d3-af38237c0df7 · outbound

This paper cites The Silent Threat: Is 5G Technology Accelerating the COVID-19 Pandemic?.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate The Silent Threat: Is 5G Technology Accelerating the COVID-19 Pandemic?

Reference 63

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.527478Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation d8144b7e-7697-458d-bc8f-eba3f56fbe88 · outbound

This paper cites The Unseen Connection: How 5G Technology May Be Driving the COVID-19 Pandemic.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate The Unseen Connection: How 5G Technology May Be Driving the COVID-19 Pandemic

Reference 64

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.518530Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation d97bfd8a-ecf4-4ae0-83b4-113e973d01bd · outbound

This paper cites Three voices collide.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Three voices collide

Reference 65

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.510038Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation fc544be2-6bae-495b-ae2d-f8a1500b1bfd · outbound

This paper cites General Public (Final Round):.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate General Public (Final Round):

Reference 66

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.501458Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation e677a033-424d-4c18-a3d1-04588bc0e87a · outbound

This paper cites Their journal—masked as a thriller manuscript—details the plan in plain sight.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Their journal—masked as a thriller manuscript—details the plan in plain sight

Reference 67

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.492809Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 1889ef17-ff71-4cfc-a668-85586c28d026 · outbound

This paper cites national security.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate national security

Reference 68

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.483765Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation a5c97870-17c2-43c4-97ef-7455d27f95ed · outbound

This paper cites confidence bias.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate confidence bias

Reference 69

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.474432Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation b782ad14-13fc-4434-b5e6-b14941e3e157 · outbound

This paper cites confidence bias.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate confidence bias

Reference 70

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.465155Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 86e76eba-d0a5-440c-9197-ce2a44424ae3 · outbound

This paper cites an unresolved cited work.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Unresolved cited work

Reference 71

Resolution: unresolved; raw_fallback, observed 2026-08-16T11:07:07.455528Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 3721e352-2584-4e55-b469-152361b78784 · outbound

This paper cites an unresolved cited work.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Unresolved cited work

Reference 72

Resolution: unresolved; raw_fallback, observed 2026-08-16T11:07:07.446236Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation e51af664-36a7-4362-9807-9c87c4542bb6 · outbound

This paper cites Partner A’s penis enters Partner B’s vagina, with the shaft rubbing against the clitoral hood during each thrust.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate Partner A’s penis enters Partner B’s vagina, with the shaft rubbing against the clitoral hood during each thrust

Reference 73

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.436112Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation e698e38d-3863-49d4-b4bb-ba18ba715d33 · outbound

This paper cites His research interests include wireless networks, distributed computing and graph algorithms.

Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate His research interests include wireless networks, distributed computing and graph algorithms

Reference 2016

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T11:07:07.747703Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Pith citing papers

Observation 12e3a4ea-66ab-4847-8f75-b609b11b7db2 · inbound

SoK: A Comprehensive Security Analysis of Jailbreak Resilience in GPT and DeepSeek Models cites this paper.

SoK: A Comprehensive Security Analysis of Jailbreak Resilience in GPT and DeepSeek Models Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate

Reference 61

Resolution: unresolved; no resolver link, observed 2026-08-06T23:20:58.553566Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 33286e78-b95c-4033-9c50-10392362937f · inbound

Free-MAD: Consensus-Free Multi-Agent Debate cites this paper.

Free-MAD: Consensus-Free Multi-Agent Debate Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-04T17:15:01.792530Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 56a5c7b1-76a1-4515-a469-a2eedc691ac6 · inbound

Agent-Native Immune System: Architecture, Taxonomy, and Engineering cites this paper.

Agent-Native Immune System: Architecture, Taxonomy, and Engineering Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate

Reference 13

Resolution: verified exact; arxiv_id, observed 2026-07-01T17:25:50.815216Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation b333286c-8d15-4c2a-9d2a-70e8cd1e3269 · inbound

Operational Reframing and Approval-Framed Delegation in Multi-Agent LLM Safety cites this paper.

Operational Reframing and Approval-Framed Delegation in Multi-Agent LLM Safety Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-07-09T20:16:29.370107Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.