arXiv AI
Jun 9
PLAGUE: Plug-and-play framework for Lifelong Adaptive Generation of Multi-turn Exploits
arXiv:2510. 17947v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are improving at an exceptional rate.
How Many Iterations to Jailbreak? Dynamic Budget Allocation for Multi-Turn LLM Evaluation
arXiv:2605. 06605v2 Announce Type: replace Abstract: Evaluating and predicting the performance of large language models (LLMs) in multi-turn conversational settings is critical yet computationally expensive; key events -- e.
Related stories
arXiv AI
Aug 10
Minimal, Local, Causal Explanations for Jailbreak Success in Large Language Models
arXiv:2605. 00123v3 Announce Type: replace Abstract: Safety trained large language models (LLMs) can often be induced to answer harmful requests through jailbreak prompts.
arXiv Machine Learning
Jun 9
Helpful to a Fault: Measuring Illicit Assistance in Multi-Turn, Multilingual LLM Agents
arXiv:2602. 16346v4 Announce Type: replace-cross Abstract: LLM-based agents execute real-world workflows via tools and memory.
arXiv AI
Jun 2
THRD: A Training-Free Multi-Turn Defense Framework for Jailbreak Attacks on Large Language Models
arXiv:2606. 01738v1 Announce Type: cross Abstract: Multi-turn jailbreak attacks pose a growing threat to LLMs by exploiting conversational dynamics such as gradual escalation and cross-turn coordination.
arXiv AI
4d ago
Does the Unsafe Gradient Survive a Conversation? On the Fragility of Gradient-Based Jailbreak Detection in Multi-Turn Dialogue
arXiv:2609.36849v1 Announce Type: cross Abstract: Safety-aligned language models are commonly deployed as multi-turn assistants, which lets adversaries spread unsafe intent across several user turns...
arXiv AI
Aug 6
Dynamic Jailbreaking Attack
arXiv:2510. 02422v4 Announce Type: replace-cross Abstract: Existing gradient-based jailbreak attacks typically optimize a fixed-length adversarial suffix toward a predefined target response with a static optimization strategy.
arXiv Machine Learning
Sep 11
Understanding In-Context Multimodal Jailbreaks via Posterior Reweighting
The paper introduces a posterior reweighting framework to explain and counter in-context learning jailbreaks in multimodal large language models. It models the model as switching between safe and harmful behavioral modes, interpreting prompt demonstrations as evidence that shifts the posterior. Using this view, the authors derive scaling laws for jailbreak effectiveness and propose a defense that injects benign counter‑evidence to suppress harmful drift while maintaining utility.
arXiv AI
Aug 20
Jailbreaking in the Haystack
The paper "Jailbreaking in the Haystack" introduces NINJA, a jailbreak technique that exploits long-context language models by appending benign, model-generated content to harmful user goals. It demonstrates that the position of harmful goals within the context is crucial for safety, and shows that NINJA significantly boosts attack success rates on models such as LLaMA, Qwen, Mistral, and Gemini. Unlike previous methods, NINJA is low-resource, transferable, less detectable, and compute‑optimal, revealing that carefully crafted benign long contexts can expose fundamental vulnerabilities in modern LMs.
arXiv AI
Aug 26
ADVERSA: Measuring Multi-Turn Guardrail Degradation and Judge Reliability in Large Language Models
The paper introduces ADVERSA, an automated red‑teaming framework that evaluates large language model safety over multiple turns by tracking continuous compliance trajectories instead of binary jailbreak outcomes. Using a fine‑tuned 70B attacker model and a structured 5‑point rubric, the authors conduct controlled experiments on three frontier victim models, measuring guardrail degradation and judge reliability through a triple‑judge consensus. Results show a 26.7% jailbreak rate with most breaches occurring early, and the study documents inter‑judge agreement, attacker drift, and attacker refusals as key factors affecting safety assessment.
arXiv AI
Jun 19
SafeSpec: Fast and Safe LLM via Dynamic Reflective Sampling
arXiv:2606. 19755v1 Announce Type: cross Abstract: Speculative inference accelerates large language model (LLM) decoding but provides no inherent safety guarantees.
arXiv Computation and Language
Sep 15
When Agents Slow Down: Understanding LLM Agents' Test-Time Strategies via Elo-per-token Analysis
arXiv:2609.15309v1 Announce Type: new Abstract: Large language model (LLM) agents allocate test-time compute adaptively as they revise solutions, use tools, explore alternatives, and decide when to s...
arXiv AI
Jun 17
In-Context Environments Induce Evaluation-Awareness in Language Models
arXiv:2603. 03824v2 Announce Type: replace Abstract: Humans often become more self-aware under threat, yet can lose self-awareness when absorbed in a task; we hypothesize that language models exhibit environment-dependent \textit{evaluation awareness}.