… Compared with strong baselines, FATE reduces attack success rate by 33.5%, harmful compliance by 82.6%, and improves external trajectory-safety diagnosis by 6.5%. These results suggest that failed trajectories can provide structured repair supervision for safer self-evolving agents. …
… The following papers were recommended by the Semantic Scholar API STARS: Skill-Triggered Audit for Request-Conditioned Invocation Safety in Agent Systems 2026 PolicyBank: Evolving Policy Understanding for LLM Agents 2026 On-Policy Self-Evolution via Failure Trajectories for Agentic Safety Alignment… …
… Rethinking Safety Evaluation for Phone-Use Agents Published on May 8 Submitted by Zhengyang Tang on May 12 Authors: Zhengyang Tang , , , , , , , , , , , , , , , , Zheng Ruan , , , , Abstract PhoneSafety benchmark reveals that avoiding harmful outcomes doesn't necessarily indicate safety, as models … …
Papers arxiv:2605.06652 When No Benchmark Exists: Validating Comparative LLM Safety Scoring Without Ground-Truth Labels Published on May 7 Submitted by Sushant Gautam on May 8 Simula Metropolitan Center for Digital Engineering SimulaMet Authors: Sushant Gautam , , , , , , , , Abstract Comparative s… …
… AI-generated summary Autonomous agents have rapidly matured as task executors and seen widespread deployment via harnesses such as OpenClaw. Safety concerns have rightly drawn growing research attention, and beneath them lie the values silently steering agent behavior. …
… We further introduce ICU -Evo to study structured-memory agents that improves long-horizon reasoning but does not fully eliminate safety failures. …
… Concern-Level Diagnostics for AI Peer Review 2026 Evaluating Patient Safety Risks in Generative AI: Development and Validation of a FMECA Framework for Generated Clinical Content 2026 PhysicianBench: Evaluating LLM Agents in Real-World EHR Environments 2026 An Empirical Study of Agent Skills for He… …
… The following papers were recommended by the Semantic Scholar API IndicSafe: A Benchmark for Evaluating Multilingual LLM Safety in South Asia 2026 ML-Bench&Guard: Policy-Grounded Multilingual Safety Benchmark and Guardrail for Large Language Models 2026 HarmChip: Evaluating Hardware Security Centri… …
… The following papers were recommended by the Semantic Scholar API SaFeR-Steer: Evolving Multi-Turn MLLMs via Synthetic Bootstrapping and Feedback Dynamics 2026 ContextualJailbreak: Evolutionary Red-Teaming via Simulated Conversational Priming 2026 Transient Turn Injection: Exposing Stateless Multi-… …
… However, their safety has received relatively limited attention. …