AI Safety

Prototype-Guided Concept Erasure in Diffusion Models CVPR 2026 Query-Routed Activation Editing with Truth-hierarchical Preference Optimization AAAI 2026 RAFFLES: Reasoning-based Attribution of Faults for LLM Systems EACL 2026 Realist and Pluralist Conceptions of Intelligence and Their Implications on AI Research AAAI 2026 Reason2Attack: Jailbreaking Text-to-Image Models via LLM Reasoning AAAI 2026 Reasoning about Uncertainty: Do Reasoning Models Know When They Don’t Know? EACL 2026 ReasoningGuard: Safeguarding Large Reasoning Models with Inference-time Safety Aha Moments ACL 2026 Reasoning or Memorization? Unreliable Results of Reinforcement Learning Due to Data Contamination AAAI 2026 Reducing Hallucinations in Language Model-based SPARQL Query Generation Using Post-Generation Memory Retrieval EACL 2026 Reference Recommendation Based Membership Inference Attack Against Hybrid-Based Recommender Systems AAAI 2026 RefusalBench: Generative Evaluation of Selective Refusal in Grounded Language Models EACL 2026 Reinforce Trustworthiness in Multimodal Emotional Support System AAAI 2026 Repairing Regex Vulnerabilities via Localization-Guided Instructions EACL 2026 Requirements for Aligned, Dynamic Resolution of Conflicts in Operational Constraints AAAI 2026 Resilience in Ambient Multi-Agent LLMs via Decentralized Bio-Autonomic Control and Immune-Inspired Anomaly Detection AAAI 2026 Response Attack: Exploiting Contextual Priming to Jailbreak Large Language Models AAAI 2026 Rethinking Deep Alignment Through the Lens of Incomplete Safety Learning AAAI 2026 Rethinking the Evaluation of Alignment Methods: Insights into Diversity, Generalisation, and Safety EACL 2026 Retrieval-Augmented Defense: Adaptive and Controllable Jailbreak Prevention for Large Language Models ACL 2026 Revisiting Unknowns: Towards Effective and Efficient Open-Set Active Learning CVPR 2026 Risk Atlas Nexus: A System for Managing AI Risks AAAI 2026 Risk-Aware Bilingual Spoken Dialogue for Campus Mental Health Support AAAI 2026 RiskLab: A Controlled Toolkit for Probing Emergent Risks in LLM-Based Multi-Agent Systems ACL 2026 Robust Adaptive Multi-Step Predictive Shielding (Student Abstract) AAAI 2026 Role-Conditioned Refusals: Evaluating Access Control Reasoning in Large Language Models EACL 2026