…a recall-safety tradeoff for clinical recommendations, and an anchoring bias to early interpretations of the patient. We further introduce ICU -Evo to study structured-memory agents that improves long-horizon reasoning…
…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 Centric LLM…
…Medical research agent skills require safeguards beyond general-purpose evaluation, including scientific integrity, methodological validity, reproducibility, and boundary safety. This study developed and preliminarily evaluated a domain-specific audit framework for medical…
…A Scalable and Diverse Multi-turn Jailbreak Benchmark for Evaluating LLM Safety (2026) Please give a thumbs up to this comment if you found it helpful! If you want recommendations for any…
…AI-generated summary Vision-language models (VLMs) have advanced rapidly and are increasingly deployed in real-world applications, especially with the rise of agent-based systems. However, their safety has received relatively…
…Geometry-Preserving Fine-Tuning for Safety in LLMs (2026) Inference-Time Vulnerability Beyond Shallow Safety: Alignment Along Generation Trajectories (2026) Different Paths to Harmful Compliance: Behavioral Side Effects and Mechanistic Divergence Across…
…Songlin Bai , Xintong Wang , , , , , , , , , , , , , Liang Ding Abstract IndustryBench evaluates industrial procurement question answering systems in Chinese against national standards, revealing significant gaps in safety compliance and highlighting the need for safety-aware…
Papers arxiv:2606.03895 Agent libOS: A Library-OS-Inspired Runtime for Long-Running, Capability-Controlled LLM Agents Published on Jun 2 Submitted by Yingqi Zhang on Jun 4 Tsinghua University Authors…
…Jona te Lintelo , , , , Abstract MASCing is a framework that enables flexible reconfiguration of Mixture-of-Experts model behavior for safety objectives through steering matrices and routing gate modifications without requiring retraining. AI…
…These design assumptions pose fundamental bottlenecks for safety in flow matching -based T2I generation, where limited sampling steps constrain iterative correction and modern context-aware text encoders diminish the effectiveness of embedding…