safety evaluation

AGENTSAFE: Benchmarking the Safety of Embodied Agents on Hazardous Instructions CVPR 2026 Among Us: Measuring and Mitigating Malicious Contributions in Model Collaboration Systems ACL 2026 Audio Jailbreak: An Open Comprehensive Benchmark for Jailbreaking Large Audio-Language Models ACL 2026 Beyond the Leaderboard: Rethinking Medical Benchmarks for Large Language Models ACL 2026 DialogGuard: Multi-Agent Psychosocial Safety Evaluation Interface of Sensitive LLM Responses ACL 2026 Do Large Language Models Reflect Demographic Pluralism in Safety? EACL 2026 MiSCHiEF: A Benchmark in Minimal-Pairs of Safety and Culture for Holistic Evaluation of Fine-Grained Image-Caption Alignment EACL 2026 MobileSafetyBench: Evaluating Safety of Autonomous Agents in Mobile Device Control AAAI 2026 Modeling Human Adversarial Strategy Adaptation in Multi-Turn Language Model Interactions CoNLL 2026 N-GLARE: An Non-Generative Latent Representation-Efficient LLM Safety Evaluator ACL 2026 Persistent Personas? Role-Playing, Instruction Following, and Safety in Extended Interactions EACL 2026 Persona-Grounded Safety Evaluation of AI Companions in Multi-Turn Conversations ACL 2026 Rethinking the Evaluation of Alignment Methods: Insights into Diversity, Generalisation, and Safety EACL 2026 RiskLab: A Controlled Toolkit for Probing Emergent Risks in LLM-Based Multi-Agent Systems ACL 2026 SDEval: Safety Dynamic Evaluation for Multimodal Large Language Models AAAI 2026 T2I-RiskyPrompt: A Benchmark for Safety Evaluation, Attack, and Defense on Text-to-Image Model AAAI 2026 TEAR: Temporal-aware Automated Red-teaming for Text-to-Video Models CVPR 2026 The Emotional Baby Is Truly Deadly: Does Your Multimodal Large Reasoning Model Have Emotional Flattery Towards Humans? AAAI 2026 USB: A COMPREHENSIVE AND UNIFIED SAFETY EVALUATION BENCHMARK FOR MULTIMODAL LARGE LANGUAGE MODELS ACL 2026 When Prompt Optimization Becomes Jailbreaking: Adaptive Red-Teaming of Large Language Models EACL 2026 A Comparative Analysis of Ethical and Safety Gaps in LLMs using Relative Danger Coefficient NAACL 2025 Adaptive Deployment of Untrusted LLMs Reduces Distributed Threats ICLR 2025 Agent-RewardBench: Towards a Unified Benchmark for Reward Modeling across Perception, Planning, and Safety in Real-World Multimodal Agents ACL 2025 AI Sandbagging: Language Models can Strategically Underperform on Evaluations ICLR 2025 Aligned LLMs Are Not Aligned Browser Agents ICLR 2025