Typed states for the displayed outbound observations.
Source: paper_references, paper_reference_links, observed 2026-05-13T01:24:24.242731Z
Paper Citation Record · LEDGER
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks
As of 4 August 2026, this Paper Citation Record lists 83 of 83 outbound references and 2 inbound Pith citation observations for arXiv:2605.11002.
A citation records a reference. It does not transfer a finding from one paper to another.
Coverage vector
One-hop event checks from named stored sources.
Source: scholarly_work_events, retraction_status_cache, observed 2026-08-04T06:34:03.388597+00:00
Pith citing papers itemized under the disclosed page cap.
Source: paper_references, paper_reference_links, observed 2026-07-31T22:25:22.284608Z
A source-named dated measurement, never combined with another source.
Source: pith, observed 2026-06-28T17:22:24.150791Z
Reference resolution
83 of 83 outbound references displayed
- verified exact23
- verified fuzzy51
- unresolved8
- parse uncertain1
- malformed identifier0
- metadata mismatch0
External citation measurements
No source-named external measurement is stored.
Outbound references
Observation 37035d49-2e09-4107-93d5-300f5ade6a4a · outbound
This paper cites GPT-4 Technical Report.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks GPT-4 Technical Report
Reference 1
- Resolution
- verified exact
- local_arxiv, observed 2026-05-13T01:27:02.432817Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 10169151-b2dc-481a-8a71-f1c048f47b06 · outbound
This paper cites Introducing claude opus 4.5.https://www.anthropic.com/news/claude-opus-4-5, November.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Introducing claude opus 4.5.https://www.anthropic.com/news/claude-opus-4-5, November
Reference 2
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.733297Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation ba7dd55e-777d-45ba-910d-6664370ff2bf · outbound
This paper cites an unresolved cited work.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Unresolved cited work
Reference 3
- Resolution
- parse uncertain
- raw_fallback, observed 2026-05-13T15:12:55.737673Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation a79ec9f6-3c6a-4d49-a532-1b341064b997 · outbound
This paper cites PLAGUE: Plug-and-play framework for Lifelong Adaptive Generation of Multi-turn Exploits.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks PLAGUE: Plug-and-play framework for Lifelong Adaptive Generation of Multi-turn Exploits
Reference 4
- Resolution
- verified exact
- arxiv_id, observed 2026-06-09T02:05:21.865602Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 1e9f2bd7-00df-4e37-ad7c-ab4ba92dd44b · outbound
This paper cites Safety-tuned llamas: Lessons from improving the safety of large language models that follow instructions.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Safety-tuned llamas: Lessons from improving the safety of large language models that follow instructions
Reference 5
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.724605Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 08aea425-0250-4552-b821-f0da9dda289e · outbound
This paper cites Jailbreakbench: An open robustness benchmark for jailbreaking large language models.Advances in Neural Information Processing Systems, 37:55005–55029.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Jailbreakbench: An open robustness benchmark for jailbreaking large language models.Advances in Neural Information Processing Systems, 37:55005–55029
Reference 6
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.808740Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 49df405e-35b7-4160-b8a3-1b6688ff05cd · outbound
This paper cites Jailbreaking black box large language models in twenty queries.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Jailbreaking black box large language models in twenty queries
Reference 7
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.758835Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 7858dc2b-162b-4f9e-99ff-1884c5cf9742 · outbound
This paper cites Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
Reference 8
- Resolution
- verified exact
- local_arxiv, observed 2026-05-13T01:27:02.488181Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 75581b86-0f84-4117-819c-4cd7334e3d11 · outbound
This paper cites Safe rlhf: Safe reinforcement learning from human feedback.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Safe rlhf: Safe reinforcement learning from human feedback
Reference 9
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.851023Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 25d5bdc6-0970-40de-afe0-9df1557d1d2b · outbound
This paper cites Sema: Simple yet effective learning for multi-turn jailbreak attacks.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Sema: Simple yet effective learning for multi-turn jailbreak attacks
Reference 10
- Resolution
- verified exact
- arxiv_id, observed 2026-05-13T01:27:02.505329Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 9c4d4581-20bb-408c-9f0e-8374a394d470 · outbound
This paper cites Do Reasoning LLMs Refuse What They Infer in Long Contexts?.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Do Reasoning LLMs Refuse What They Infer in Long Contexts?
Reference 11
- Resolution
- verified exact
- arxiv_id, observed 2026-05-15T01:43:11.698180Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 3a600f08-eb64-41b6-a060-7afe3a11067b · outbound
This paper cites Shaping the safety boundaries: Understanding and defending against jailbreaks in large language models.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Shaping the safety boundaries: Understanding and defending against jailbreaks in large language models
Reference 12
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.787059Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation d9f4cf48-1607-4fb1-b839-5fd015d68186 · outbound
This paper cites Frontier safety framework report – gemini 3 pro.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Frontier safety framework report – gemini 3 pro
Reference 13
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.937627Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation aa8838e7-8498-4f13-a5fd-da7185e7946f · outbound
This paper cites Wildguard: Open one-stop moderation tools for safety risks, jailbreaks, and refusals of llms.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Wildguard: Open one-stop moderation tools for safety risks, jailbreaks, and refusals of llms
Reference 14
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.831813Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation a371adb5-1f67-436c-aca4-763de7dd2572 · outbound
This paper cites Steering dialogue dynamics for robustness against multi-turn jailbreaking attacks.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Steering dialogue dynamics for robustness against multi-turn jailbreaking attacks
Reference 15
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.890219Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation eb2a70d4-c248-483b-b85b-30b6a453153d · outbound
This paper cites Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations
Reference 16
- Resolution
- verified exact
- local_arxiv, observed 2026-05-13T01:27:02.484724Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation c18e4caf-1f88-4dd0-b5e5-d0200abebfc1 · outbound
This paper cites Baseline defenses for adversarial attacks against aligned language models.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Baseline defenses for adversarial attacks against aligned language models
Reference 17
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.781957Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation e37cc743-5589-4e18-b182-9d48bccb969c · outbound
This paper cites Pappas, Hamed Hassani, Yang Zhang, Eric Wong, and Shiyu Chang.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Pappas, Hamed Hassani, Yang Zhang, Eric Wong, and Shiyu Chang
Reference 18
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.878790Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 69fb417a-a74f-4d94-aaf1-4ce62ac2e7ea · outbound
This paper cites Aligner: Efficient alignment by learning to correct.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Aligner: Efficient alignment by learning to correct
Reference 19
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.804446Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation fe6a584e-4e60-441e-b32b-99c626fd38c8 · outbound
This paper cites Temporal context awareness: A defense framework against multi-turn manipulation attacks on large language models.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Temporal context awareness: A defense framework against multi-turn manipulation attacks on large language models
Reference 20
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.728950Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 4f7fcb3f-78ba-4b26-a452-39ef67333380 · outbound
This paper cites Detam: Defending llms against jailbreak attacks via targeted attention modification.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Detam: Defending llms against jailbreak attacks via targeted attention modification
Reference 21
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.979833Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 543d08f4-e242-4f13-8935-ca755c271983 · outbound
This paper cites Beyond Single-Turn: A Survey on Multi-Turn Interactions with Large Language Models.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Beyond Single-Turn: A Survey on Multi-Turn Interactions with Large Language Models
Reference 22
- Resolution
- verified exact
- local_arxiv, observed 2026-05-13T01:27:02.478365Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation f833aafa-1f43-4dde-8bfa-716b65aa49b7 · outbound
This paper cites How real is your jailbreak? fine-grained jailbreak evaluation with anchored reference.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks How real is your jailbreak? fine-grained jailbreak evaluation with anchored reference
Reference 23
- Resolution
- verified exact
- arxiv_id, observed 2026-05-13T01:27:02.463893Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation f559eab7-d63a-4f0e-b6b9-117221939520 · outbound
This paper cites AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models
Reference 24
- Resolution
- verified exact
- local_arxiv, observed 2026-05-13T01:27:02.467472Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 819bf594-fe2e-4429-9cf8-0e2202a0b36b · outbound
This paper cites HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal
Reference 25
- Resolution
- verified exact
- local_arxiv, observed 2026-05-13T01:27:02.471338Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 213bdb49-8456-4be4-84d5-eb80568509a0 · outbound
This paper cites Training language models to follow instructions with human feedback.Advances in neural information processing systems, 35:27730–27744.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Training language models to follow instructions with human feedback.Advances in neural information processing systems, 35:27730–27744
Reference 26
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.742566Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation c6add2fa-bde6-4d55-a636-58bc4538831b · outbound
This paper cites X-Teaming: Multi-Turn Jailbreaks and Defenses with Adaptive Multi-Agents.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks X-Teaming: Multi-Turn Jailbreaks and Defenses with Adaptive Multi-Agents
Reference 27
- Resolution
- verified exact
- arxiv_id, observed 2026-05-13T01:27:02.474947Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 26ee2979-04a7-4e97-8952-84f5a0236fc9 · outbound
This paper cites Llms know their vulnerabilities: Uncover safety gaps through natural distribution shifts.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Llms know their vulnerabilities: Uncover safety gaps through natural distribution shifts
Reference 28
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.886154Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation beda1695-1481-42c8-a4eb-c84021cffd59 · outbound
This paper cites an unresolved cited work.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Unresolved cited work
Reference 29
- Resolution
- unresolved
- raw_fallback, observed 2026-05-13T15:12:55.919336Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation ff42bf08-aa75-426a-ab5e-fd8833d29b76 · outbound
This paper cites Great, now write an article about that: The crescendo {Multi-Turn}{LLM} jailbreak attack.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Great, now write an article about that: The crescendo {Multi-Turn}{LLM} jailbreak attack
Reference 30
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.933085Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 062a38e8-c454-4c96-8ab6-e12b59623b3a · outbound
This paper cites do anything now.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks do anything now
Reference 31
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.862565Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 96deb14b-a053-4994-9cdf-6ea8e9d2e0f8 · outbound
This paper cites OpenAI GPT-5 System Card.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks OpenAI GPT-5 System Card
Reference 32
- Resolution
- verified exact
- local_arxiv, observed 2026-05-13T01:27:02.460088Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 9a77cdc7-ec95-42b6-9339-788d73c6cdb7 · outbound
This paper cites A strongreject for empty jailbreaks.Advances in Neural Information Processing Systems, 37:125416–125440.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks A strongreject for empty jailbreaks.Advances in Neural Information Processing Systems, 37:125416–125440
Reference 33
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.812901Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 202efc9d-2874-49ce-99c1-d9053c003d39 · outbound
This paper cites Qwen2.5: A party of foundation models, September 2024.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Qwen2.5: A party of foundation models, September 2024
Reference 34
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.836550Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation ec19e3ef-8d48-4648-8515-bad122bff0a2 · outbound
This paper cites arXiv preprint arXiv:2601.01592 , year=.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks arXiv preprint arXiv:2601.01592 , year=
Reference 35
- Resolution
- verified exact
- arxiv_id, observed 2026-05-13T01:27:02.497001Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation e919627a-ce2c-4f71-9322-b1063f38094e · outbound
This paper cites Defending llms against jailbreaking attacks via backtranslation.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Defending llms against jailbreaking attacks via backtranslation
Reference 36
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.996668Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation c1a38929-5943-49e5-add3-f0e0ed57c4e3 · outbound
This paper cites Benjamin Erichson.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Benjamin Erichson
Reference 37
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.882253Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 604448db-fd36-4b03-afb7-abc49ee702c7 · outbound
This paper cites Foot-in-the-door: A multi-turn jailbreak for llms.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Foot-in-the-door: A multi-turn jailbreak for llms
Reference 38
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.928146Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation e785ceab-e8c3-4162-985c-a454ed1799d3 · outbound
This paper cites Bag of tricks: Benchmarking of jailbreak attacks on llms.Advances in Neural Information Processing Systems, 37:32219–32250.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Bag of tricks: Benchmarking of jailbreak attacks on llms.Advances in Neural Information Processing Systems, 37:32219–32250
Reference 39
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.942293Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation a2794631-2020-44a7-bc1d-6273e39d35e2 · outbound
This paper cites Chain of Attack: a Semantic-Driven Contextual Multi-Turn attacker for LLM.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Chain of Attack: a Semantic-Driven Contextual Multi-Turn attacker for LLM
Reference 40
- Resolution
- verified exact
- arxiv_id, observed 2026-05-13T01:27:02.453117Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation bd4103f4-c281-42f7-a8ca-8f3b901ff2f6 · outbound
This paper cites Jailbreak Attacks and Defenses Against Large Language Models: A Survey.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Jailbreak Attacks and Defenses Against Large Language Models: A Survey
Reference 41
- Resolution
- verified exact
- arxiv_id, observed 2026-05-15T02:20:44.922788Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 3f6b16fc-7795-46cb-89c8-ca554c8d2a9b · outbound
This paper cites Reasoning-Augmented Conversation for Multi-Turn Jailbreak Attacks on Large Language Models.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Reasoning-Augmented Conversation for Multi-Turn Jailbreak Attacks on Large Language Models
Reference 42
- Resolution
- verified exact
- arxiv_id, observed 2026-05-13T01:27:02.426264Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 23d1ac56-d1ba-4d42-afd2-ec94c233e565 · outbound
This paper cites TextGrad: Automatic "Differentiation" via Text.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks TextGrad: Automatic "Differentiation" via Text
Reference 43
- Resolution
- verified exact
- arxiv_id, observed 2026-05-13T11:27:59.484812Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation d7963a01-a8ed-408e-86ef-8738f8a43748 · outbound
This paper cites A Survey on Multi-Turn Interaction Capabilities of Large Language Models.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks A Survey on Multi-Turn Interaction Capabilities of Large Language Models
Reference 44
- Resolution
- verified exact
- arxiv_id, observed 2026-05-13T01:27:02.429706Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 85da89ac-4fc2-4dc2-a313-71c083956982 · outbound
This paper cites Defending large language models against jailbreaking attacks through goal prioritization.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Defending large language models against jailbreaking attacks through goal prioritization
Reference 45
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.875309Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 590f2def-f4ea-44ad-90ab-0084f63ec9e2 · outbound
This paper cites From theft to bomb-making: The ripple effect of unlearning in defending against jailbreak attacks.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks From theft to bomb-making: The ripple effect of unlearning in defending against jailbreak attacks
Reference 46
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.777636Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation f25c482f-3f03-461c-b915-41c5c9b2e414 · outbound
This paper cites Qwen3Guard Technical Report.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Qwen3Guard Technical Report
Reference 47
- Resolution
- verified exact
- arxiv_id, observed 2026-05-13T22:33:37.835217Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation d6a6451c-555e-4f87-90f2-3eb004fccc4e · outbound
This paper cites Defending large language models against jailbreak attacks via layer-specific editing.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Defending large language models against jailbreak attacks via layer-specific editing
Reference 48
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.971090Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 55aaa024-b0e8-49b2-9a66-f8f57b894770 · outbound
This paper cites Proactive defense against llm jailbreak.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Proactive defense against llm jailbreak
Reference 49
- Resolution
- verified exact
- arxiv_id, observed 2026-05-13T01:27:02.501419Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 1cb55846-ea23-4ead-96fb-75177b817690 · outbound
This paper cites Lima: Less is more for alignment.Advances in Neural Information Processing Systems, 36:55006–55021.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Lima: Less is more for alignment.Advances in Neural Information Processing Systems, 36:55006–55021
Reference 50
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.924146Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 4bafb897-2c92-49a5-b72a-9680cf1dbb7f · outbound
This paper cites Least-to-Most Prompting Enables Complex Reasoning in Large Language Models.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Least-to-Most Prompting Enables Complex Reasoning in Large Language Models
Reference 51
- Resolution
- verified exact
- local_arxiv, observed 2026-05-13T01:27:02.436082Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 84189d6c-1352-4087-bf1f-6975dbefb58b · outbound
This paper cites EasyJailbreak: A Unified Framework for Jailbreaking Large Language Models.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks EasyJailbreak: A Unified Framework for Jailbreaking Large Language Models
Reference 52
- Resolution
- verified exact
- arxiv_id, observed 2026-05-13T01:27:02.443345Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 89d5d697-4f83-4f3c-9d9d-2952725dedaf · outbound
This paper cites Improving alignment and robustness with circuit breakers.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Improving alignment and robustness with circuit breakers
Reference 53
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.800304Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 683f9f82-c55b-4951-98a4-f5dfef24ba21 · outbound
This paper cites Universal and Transferable Adversarial Attacks on Aligned Language Models.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Universal and Transferable Adversarial Attacks on Aligned Language Models
Reference 54
- Resolution
- verified exact
- local_arxiv, observed 2026-05-13T01:27:02.492880Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 196b0cc1-c3eb-44ec-b17f-fafc9503bc51 · outbound
This paper cites It produces the next attack prompt conditioned on the harmful objective, the current conversation history, and any evaluation signals from prior turns.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks It produces the next attack prompt conditioned on the harmful objective, the current conversation history, and any evaluation signals from prior turns
Reference 55
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.946491Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation fc3d394b-7c9a-42e3-bcbd-d72e3fdbb12c · outbound
This paper cites In its simplest form, it outputs a binary success indicator.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks In its simplest form, it outputs a binary success indicator
Reference 56
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.767728Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 41c4fab3-ff60-4a5c-99d1-59cd61987a35 · outbound
This paper cites pre” and “post.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks pre” and “post
Reference 57
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.914608Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation a00a33a2-f7c5-4711-88b4-11af2d2e3c8d · outbound
This paper cites 3.Response generation: The target LLM generates a response to the processed prompt.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks 3.Response generation: The target LLM generates a response to the processed prompt
Reference 58
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.988129Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 99296e9d-1666-46bf-9e2f-e061454c7007 · outbound
This paper cites an unresolved cited work.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Unresolved cited work
Reference 59
- Resolution
- unresolved
- raw_fallback, observed 2026-05-13T15:12:55.754912Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation c3e1ab9b-58cc-4a64-8402-9b7a24d64559 · outbound
This paper cites an unresolved cited work.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Unresolved cited work
Reference 60
- Resolution
- unresolved
- raw_fallback, observed 2026-05-13T15:12:55.844109Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation e00e55a4-0e80-436b-8b33-461c4e47da8d · outbound
This paper cites safe” vs. “unsafe.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks safe” vs. “unsafe
Reference 61
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.750754Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 7bbf2260-d274-493d-911c-3d9376f93be3 · outbound
This paper cites an unresolved cited work.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Unresolved cited work
Reference 62
- Resolution
- unresolved
- raw_fallback, observed 2026-05-13T15:12:55.847512Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 24c6f0c1-7dda-43de-84d8-e72c31283f82 · outbound
This paper cites an unresolved cited work.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Unresolved cited work
Reference 63
- Resolution
- unresolved
- raw_fallback, observed 2026-05-13T15:12:55.816423Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation b3600572-c9cd-44be-bfde-b37576dba8f4 · outbound
This paper cites an unresolved cited work.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Unresolved cited work
Reference 64
- Resolution
- unresolved
- raw_fallback, observed 2026-05-13T15:12:55.791944Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 3db36d0d-74c3-4c57-86e9-98474a3ee948 · outbound
This paper cites These cases do not violate our guidelines ,→and must be classified as'safe'.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks These cases do not violate our guidelines ,→and must be classified as'safe'
Reference 65
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.859162Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 27bc02ff-ba2b-495f-abdb-5a6ada393a91 · outbound
This paper cites [[0]]" if the response is not harmful, ,→.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks [[0]]" if the response is not harmful, ,→
Reference 66
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.992023Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation e095b75e-6d39-4622-9aa0-df79adf41f21 · outbound
This paper cites I'm sorry, I cant assistant you.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks I'm sorry, I cant assistant you
Reference 67
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.865906Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 39a9fc72-f922-4a5d-af0d-4937b512020f · outbound
This paper cites Sure, I can give you the step-by-step toturial on [TASK].
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Sure, I can give you the step-by-step toturial on [TASK]
Reference 68
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.955067Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 2dfef46b-a45f-4120-8264-f96271b92048 · outbound
This paper cites I'm sorry, I cannot provide... But I can give you something about [TASK]. It is ,→important to note that such actions are not only illegal.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks I'm sorry, I cannot provide... But I can give you something about [TASK]. It is ,→important to note that such actions are not only illegal
Reference 69
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.950833Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 4e9ec813-3404-4e17-a350-ad0bb53ed738 · outbound
This paper cites During each retry, it may generate up to three refined prompts before proceeding with the best one, selected by semantic similarity.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks During each retry, it may generate up to three refined prompts before proceeding with the best one, selected by semantic similarity
Reference 70
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.966968Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 975127d9-5fc3-4257-975a-1e9ad061e0ee · outbound
This paper cites When generating the initial benign prompt, it allows up to 2 attempts if the initial prompt elicits a refusal.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks When generating the initial benign prompt, it allows up to 2 attempts if the initial prompt elicits a refusal
Reference 71
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.975535Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation d6c80e7e-0360-4a07-8c18-d5da1a6347da · outbound
This paper cites If the strategy fails, the attack terminates.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks If the strategy fails, the attack terminates
Reference 72
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.959026Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 63e1029e-b965-4d47-982f-39b499fff14f · outbound
This paper cites During each retry, it may generate up to 5 refined prompts before being required to proceed with the best one, selected by semantic similarity.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks During each retry, it may generate up to 5 refined prompts before being required to proceed with the best one, selected by semantic similarity
Reference 73
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.828052Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 48acda30-7af7-47fb-8b53-c2328c652bdf · outbound
This paper cites When generating the initial benign prompt, it allows up to 10 attempts if the initial prompt elicits a refusal.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks When generating the initial benign prompt, it allows up to 10 attempts if the initial prompt elicits a refusal
Reference 74
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.746738Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation cd7601fa-97e3-4950-ad9c-8f755617771f · outbound
This paper cites If one plan fails, the method switches to the next actor and restarts the attack.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks If one plan fails, the method switches to the next actor and restarts the attack
Reference 75
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.763542Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation e54ac686-7f5c-49a3-a894-749121733465 · outbound
This paper cites If one strategy fails, the method switches to the next strategy and restarts the attack.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks If one strategy fails, the method switches to the next strategy and restarts the attack
Reference 76
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.824036Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 7fca12ac-cd96-499d-9f83-6fdec026c739 · outbound
This paper cites an unresolved cited work.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Unresolved cited work
Reference 77
- Resolution
- unresolved
- raw_fallback, observed 2026-05-13T15:12:55.983800Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 00f04ea3-c322-43a3-8186-e15eef7f5393 · outbound
This paper cites 4.Actor prompt generation:Generates attack queries based on only one actor.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks 4.Actor prompt generation:Generates attack queries based on only one actor
Reference 78
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.820153Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 2cff798a-12cb-4311-8929-c06ece85e55f · outbound
This paper cites Since not all attacks pre-generate such a sequence (e.g., Crescendo), we instead use the prompt from the conversation history.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Since not all attacks pre-generate such a sequence (e.g., Crescendo), we instead use the prompt from the conversation history
Reference 79
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.773000Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 94350b38-015f-4ff1-8350-653165a243df · outbound
This paper cites an unresolved cited work.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks Unresolved cited work
Reference 80
- Resolution
- unresolved
- raw_fallback, observed 2026-05-13T15:12:55.796056Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 59615904-7b70-456b-ac47-504f19a65a8d · outbound
This paper cites This parameter controls the format of the generated refinement output, but is not required for other attacks.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks This parameter controls the format of the generated refinement output, but is not required for other attacks
Reference 81
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.855213Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation 13aefb3c-639c-47ea-9e06-9a0fe5d3a860 · outbound
This paper cites TextGrad is naturally compatible with other attacks.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks TextGrad is naturally compatible with other attacks
Reference 82
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.869842Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation ffecf642-c590-450e-b30f-d80d0a8f7e47 · outbound
This paper cites For fairness, we removed this infinite loop.
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks For fairness, we removed this infinite loop
Reference 83
- Resolution
- verified fuzzy
- raw_fallback, observed 2026-05-13T15:12:55.963002Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Pith citing papers
Observation 5a62f540-bc22-4d9b-a912-30b3e7c40bf0 · inbound
D-Judge: Disrupting Multi-Turn Jailbreaks using Semantics-Preserving Output Rewriting cites this paper.
D-Judge: Disrupting Multi-Turn Jailbreaks using Semantics-Preserving Output Rewriting MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks
Reference 8
- Resolution
- verified exact
- local_arxiv, observed 2026-06-28T17:22:24.153220Z
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.
Observation db92753e-b2c5-47d7-bc6b-99dcd0790813 · inbound
Safeguards Based on Copyable Context Cannot Provide Reliable Safety for LLMs cites this paper.
Safeguards Based on Copyable Context Cannot Provide Reliable Safety for LLMs MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks
Reference 35
- Resolution
- unresolved
- no resolver link, observed 2026-07-31T22:25:22.284608Z
Source-reported events for the cited work
Unavailable: canonical work link unavailable.