arXiv Computation and Language

Evaluating Language Model Safety Across Long Adversarial Conversations

The paper investigates how conversational safety in language models degrades over extended, adversarial interactions. By testing three instruction‑tuned models with persistent adversarial users across up to 101 turns, the study finds that safe‑response rates drop sharply from 85–100% at the first turn to 15–44% by the end. This demonstrates that strong single‑turn safety does not guarantee continued safety in long conversations.

arXiv AI

Sep 24

Beyond Unsafe Detection: Counterfactually Anchored Evidence Attribution for Multi-Turn LLM Safety Failures

The paper introduces a counterfactually anchored evidence attribution approach for multi‑turn large language model safety failures. It presents a new dataset of 1,762 conversations, including adversarial, benign twins, and high‑risk vocabulary variants, and trains a lightweight hierarchical model that accurately predicts safety violations and attributes them to specific user turns and token spans. The model achieves high detection performance (F1 = 0.988) and significantly reduces adversarial confidence when top‑attributed tokens are removed, while maintaining low false‑positive rates on benign conversations.

By Srinivasan Subramanian, Kazi Aminul Islam, Md. Abdullah Al Hafiz Khan

llms agents safety

arXiv AI

Jul 22

A Red Teaming Framework for Large Language Models: A Case Study on Faithfulness Evaluation

arXiv:2606. 25476v2 Announce Type: replace-cross Abstract: Large language models (LLMs) have demonstrated remarkable performance across natural language processing tasks, yet their deployment in high-stakes applications raises critical concerns regarding reliability, safety, and trustworthiness.

By Abrar Alotaibi, Raed Mughus, Moataz Ahmed

llms nlp safety

arXiv Computation and Language

3d ago

The Geometry of Harmfulness in Multi-Turn Attacks

arXiv:2609.38389v1 Announce Type: cross Abstract: Large language models (LLMs) remain vulnerable to adversarial attacks that circumvent safety alignment to elicit harmful outputs. It remains unclear...

By Yelyzaveta (Lisa), Husieva, Lauren Alvarez

llms safety

arXiv AI

Aug 6

Temporal Context Awareness: A Defense Framework Against Multi-turn Manipulation Attacks on Large Language Models

arXiv:2503. 15560v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly vulnerable to sophisticated multi-turn manipulation attacks, where adversaries strategically build context through seemingly benign conversational turns to circumvent safety measures and elicit harmful or unauthorized responses.

By Prashant Kulkarni, Assaf Namer

llms rag robotics safety

arXiv AI

Jul 24

Robust Critics: Defending LLMs Against Multi-Turn Attacks

arXiv:2607. 20472v1 Announce Type: new Abstract: When a user asks a language model something harmful, is it a genuine attack or a misunderstood but well-meaning question?

By Roman Belaire, Arunesh Sinha, Pradeep Varakantham

llms reinforcement-learning fine-tuning safety

arXiv Machine Learning

Jun 2

Investigating and Alleviating Harm Amplification in LLM Interactions

arXiv:2606. 02423v1 Announce Type: cross Abstract: Large language models (LLMs) can serve as helpful assistants, yet they can equally function as harm amplifiers that enable malicious users to achieve harmful outcomes beyond their capabilities through extended interactions.

By Ruohao Guo, Wei Xu, Alan Ritter

llms benchmarks safety

Hugging Face Trending Papers

Jun 4

Safety Paradox: How Enhanced Safety Awareness Leaves LLMs Vulnerable to Posterior Attack

Large language models (LLMs) are rigorously aligned to refuse harmful requests, a process that inherently cultivates a latent capacity to evaluate and recognize unsafe content. In this work, we reveal that this advanced safety awareness inadvertently introduces a fatal vulnerability.

llms reinforcement-learning safety

arXiv Computation and Language

Aug 25

No One Model Catches Every Harm: Benchmarking Content Moderation Across Safety Scenarios

arXiv:2608.21775v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly deployed in real-world applications, yet they remain vulnerable to generating harmful content. From adver...

By Afshin Orojlooyjadid, Hitesh Patel

llms benchmarks safety

arXiv AI

Aug 24

Certified Multi-Turn Robustness for LLM Safety via Compositional Bounds and Safety Persistence

arXiv:2608. 20820v1 Announce Type: new Abstract: Large language models (LLMs) are vulnerable to multi-turn jailbreak attacks that progressively manipulate conversation context.

By Yang Liu, Bin Chong, Wenkai Yang, Shuai Zhang, Yancheng Chen, Feiyu Han, GuoZhen, Cheng Zhang, Huaibing Xie, Changze Lv, Shihan Dou, Pluto Zhou

llms rag safety

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...

By Omar Sheta, Rinku Deuja, Hadi Masoudi, Minghong Fang

llms safety

arXiv Computation and Language

Sep 4

Towards Multi-modal Multi-turn Safety: From Agentic Interaction to Strategic Alignment

The paper introduces MINT‑Safe, a new open‑source dataset of 11,270 multi‑image dialogues and 500 refusal VQA pairs designed to expose safety risks in multi‑modal large language models during open‑ended conversations. It also proposes TAD‑Align, a turn‑aware dual‑objective reward framework that dynamically up‑weights dialogue turns with inconsistent safety behavior, improving safety metrics on Qwen2.5‑VL‑7B‑Instruct and LLaVA‑Next‑7B. The results show over 10% reduction in attack success rate and notable gains in harmlessness and helpfulness while maintaining overall model performance.

By Han Zhu, Jiale Chen, Chengkun Cai, Shengjie Sun, Haoran Li, Yujin Zhou, Chi-Min Chan, Pengcheng Wen, Lei Li, Yike Guo, Sirui Han

llms agents diffusion reinforcement-learning multimodal benchmarks safety

arXiv AI

Jul 31

State-Dependent Safety Failures in Multi-Turn Language Model Interaction

arXiv:2603. 15684v2 Announce Type: replace-cross Abstract: Safety alignment in large language models is typically evaluated under isolated queries, yet real-world use is inherently multi-turn.

By Pengcheng Li, Jie Zhang, Tianwei Zhang, Han Qiu, Zhang kejun, Weiming Zhang, Nenghai Yu, Wenbo Zhou

llms safety

Related stories