Pith. sign in

REVIEW 9 cited by

X-Teaming: Multi-Turn Jailbreaks and Defenses with Adaptive Multi-Agents

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2504.13203 v2 pith:ZXP2FNIO submitted 2025-04-15 cs.CR cs.AIcs.CLcs.LGcs.MA

classification cs.CRcs.AIcs.CLcs.LGcs.MA
keywords multi-turnsafetyx-teamingattackacrossattackschallengesdiversity
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Multi-turn interactions with language models (LMs) pose critical safety risks, as harmful intent can be strategically spread across exchanges. Yet, the vast majority of prior work has focused on single-turn safety, while adaptability and diversity remain among the key challenges of multi-turn red-teaming. To address these challenges, we present X-Teaming, a scalable framework that systematically explores how seemingly harmless interactions escalate into harmful outcomes and generates corresponding attack scenarios. X-Teaming employs collaborative agents for planning, attack optimization, and verification, achieving state-of-the-art multi-turn jailbreak effectiveness and diversity with success rates up to 98.1% across representative leading open-weight and closed-source models. In particular, X-Teaming achieves a 96.2% attack success rate against the latest Claude 3.7 Sonnet model, which has been considered nearly immune to single-turn attacks. Building on X-Teaming, we introduce XGuard-Train, an open-source multi-turn safety training dataset that is 20x larger than the previous best resource, comprising 30K interactive jailbreaks, designed to enable robust multi-turn safety alignment for LMs. Our work offers essential tools and insights for mitigating sophisticated conversational attacks, advancing the multi-turn safety of LMs.

Discussion (0). Sign in to comment.

Forward citations

Cited by 9 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. SoK: Intent-Oriented Systematization of Multi-Turn LLM Jailbreaks

    cs.CR 2026-08 conditional novelty 7.0 of 10

    Multi-turn LLM jailbreaks succeed based on how harmful intent is organized across turns, not on interaction length, and detection should shift to session and cross-session scope.

  2. Do LLMs Know Their Vulnerable Scenarios?

    cs.AI 2026-07 conditional novelty 7.0 of 10

    Scenario jailbreaks suppress refusal via internal concept directions; Concept2Scenario attributes those concepts with SAEs and turns them into transferable natural-language attack scenarios.

  3. AMT-X: Phase-Structured Multi-Turn Red-Teaming with Checklist-Gated Evaluation

    cs.CR 2026-07 conditional novelty 6.0 of 10

    A phase-structured multi-turn red-team framework reports 97.6–100% lenient ASR but only 66.7–78.6% full actionable ASR on six frontier LLMs, with success strongly depth-dependent.

  4. Robust Critics: Defending LLMs Against Multi-Turn Attacks

    cs.AI 2026-05 conditional novelty 6.0 of 10

    Critic-weighted sampling over inferred user intents improves multi-turn LLM defense success while preserving helpfulness, with an expected-Q improvement guarantee and transfer to frontier models.

  5. Reasoning Up the Instruction Ladder for Controllable Language Models

    cs.CL 2025-10 conditional novelty 6.0 of 10

    RLVR on ~7K verifiable system/user conflict examples teaches LLMs to prioritize higher-priority instructions, improving instruction-hierarchy and safety benchmarks.

  6. SafeWork-R1: Coevolving Safety and Intelligence under the AI-45$^{\circ}$ Law

    cs.AI 2025-07 conditional novelty 6.0 of 10

    SafeWork-R1 shows that a staged RL pipeline with safety, value, and knowledge verifiers can improve both safety and general reasoning scores over a base multimodal model.

  7. Automatic Hard Example Synthesis with Multi-Level Agentic Data Curation

    cs.AI 2026-07 conditional novelty 5.0 of 10

    Test-time retrieval of committee-disagreement-mined synthetic images cuts a safety classifier's false-negative rate on a hard HoliSafe subset from 41.2% to 24.5%.

  8. A Multimodal Automatic Redteaming Evaluation based on Atomic Jailbreak Strategy Decoupling and Combination

    cs.CR 2026-08 conditional novelty 4.0 of 10

    A new jailbreak framework, HACA, combines atomic text and image attack strategies selected by a cross-modal planner and generates attacks with LLMs and text-to-image models, reaching 95.48% average attack success acro...

  9. Stateful Cooperative Agents Safeguarding LLMs Against Evolving Multi-Turn Attacks

    cs.CR 2026-07 reject novelty 4.0 of 10

    CoopGuard's defer-tempt-analyze-coordinate agents cut reported jailbreak success and raise attacker token costs on the new EMRA benchmark, but the deceptive-rate metric is partly defined by the paper's own scoring rubric.

Pith tools