AI Safety

Quantifying Misalignment Between Agents: Towards a Sociotechnical Understanding of Alignment AAAI 2025 Quantifying the Influence of Irrelevant Contexts on Political Opinions Produced by LLMs ACL 2025 Quantitative Predictive Monitoring and Control for Safe Human-Machine Interaction AAAI 2025 QueryAttack: Jailbreaking Aligned Large Language Models Using Structured Non-natural Query Language ACL 2025 RAG LLMs are Not Safer: A Safety Analysis of Retrieval-Augmented Generation for Large Language Models NAACL 2025 Rainbow-Teaming for the Polish Language: A Reproducibility Study NAACL 2025 Random-Set Neural Networks ICLR 2025 Rapid Mismatch Estimation via Neural Network Informed Variational Inference CoRL 2025 RealSafe: Quantifying Safety Risks of Language Agents in Real-World COLING 2025 Real-Time Out-of-Distribution Failure Prevention via Multi-Modal Reasoning CoRL 2025 Reasoning-Augmented Conversation for Multi-Turn Jailbreak Attacks on Large Language Models EMNLP 2025 Reasoning Model Unlearning: Forgetting Traces, Not Just Answers, While Preserving Reasoning Skills EMNLP 2025 Reasoning-to-Defend: Safety-Aware Reasoning Can Defend Large Language Models from Jailbreaking EMNLP 2025 RE-Bench: Evaluating Frontier AI R&D Capabilities of Language Model Agents against Human Experts ICML 2025 RedactOR: An LLM-Powered Framework for Automatic Clinical Data De-Identification ACL 2025 RedHit: Adaptive Red-Teaming of Large Language Models via Search, Reasoning, and Preference Optimization ACL 2025 Red Queen: Exposing Latent Multi-Turn Risks in Large Language Models ACL 2025 Red-Teaming for Uncovering Societal Bias in Large Language Models ACL 2025 Red-Teaming LLM Multi-Agent Systems via Communication Attacks ACL 2025 Reducing Tool Hallucination via Reliability Alignment ICML 2025 RED: Unleashing Token-Level Rewards from Holistic Feedback via Reward Redistribution EMNLP 2025 REFINE: Inversion-Free Backdoor Defense via Model Reprogramming ICLR 2025 Reflective Verbal Reward Design for Pluralistic Alignment IJCAI 2025 Refusal-Aware Red Teaming: Exposing Inconsistency in Safety Evaluations EMNLP 2025 Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Training ACL 2025