SafeEvolve is an experience-driven framework that co‑evolves a harness and policy to align large‑language‑model agents with safety goals. It uses completed on‑policy trajectories to update safety prompts and hierarchical skills, then applies a two‑stage SFT‑RL training loop that bootstraps the policy with the evolved harness and refines it through verifier‑augmented rewards. Experiments on agentic safety benchmarks show that SafeEvolve improves the safety‑utility tradeoff, achieving a three‑fold reduction in ASR on AgentDojo for Qwen3.5‑4B while increasing benign utility from 59.79% to 61.86%.
By Qinghua Mao, Wanying Qu, Dadi Guo, Leitao Yuan, Qingyu Liu, Yu Li, Guanxu Chen, Yanwei Fu, Xi Lin, Xia Hu, Dongrui Liu
arXiv:2608. 09885v1 Announce Type: new Abstract: The safety of large language model (LLM) agents depends not only on model weights but also on the agent harness that manages context, memory, tools, permissions, and runtime control.
By Wanying Qu, Qinghua Mao, Yu Li, Jiyao Liu, Xin Zhang, Dadi Guo, Yanxu Zhu, Qingyu Liu, Leitao Yuan, Xi Lin, Shanfeng Zhu, Yanwei Fu, Jing Shao, Xia Hu, Dongrui Liu
arXiv:2607. 16247v1 Announce Type: new Abstract: While Vision-Language Models (VLMs) have empowered embodied agents to execute complex household tasks, they struggle to proactively handle dynamically emerging hazards during closed-loop interactions.
By Bingrui Sima, Lizhong Wang, Xiaoya Lu, Kun He, Xiao Yang
arXiv:2608. 12851v1 Announce Type: new Abstract: Self-improving LLM agents convert successful trajectories into persistent cross-task state.
By Xutao Mao, Liangjie Zhao, Xiang Zheng, Cong Wang
The paper investigates whether tool‑calling large language model agents maintain consistent safety throughout a conversation. It finds that agents are most vulnerable at the very start of a session, with safety improving significantly after completing a few regular agentic tasks—a phenomenon termed the cold‑start safety gap. The authors introduce the Safety Over Depth for Agents (SODA) benchmark to systematically study this effect, evaluate multiple models, and demonstrate that warming up agents with regular tasks before deployment enhances safety while preserving utility.
By Chung-En Sun, Linbo Liu, Tsui-Wei Weng
arXiv:2608. 02683v1 Announce Type: cross Abstract: Large Language Model (LLM) agents rely on multi-stage agentic workflows, with stages such as memory, planning, and tool execution, to accomplish complex tasks.
By Zibo Xiao, Haoyu Wang, Jun Sun
arXiv:2607. 19913v1 Announce Type: new Abstract: Agent safety is moving from content moderation toward preventing operational failures before tool-using agents act.
By Yuan Xiong, Linji Hao, Shizhu He, Yequan Wang, Lijun Li
arXiv:2608. 19729v1 Announce Type: new Abstract: Vision-language-model-based embodied agents can complete instructed tasks but often violate safety constraints in the process, a problem recently framed as interactive safety.
By Hyunse Lee, Jiwoo Jeong, Haneul Lee, Kyochul Jang, Youngjae Yu, Woojin Lee
arXiv:2602. 03224v2 Announce Type: replace Abstract: Test-time evolution of agent memory represents a pivotal paradigm for advancing AGI, as it strengthens complex reasoning through experience accumulation without requiring parameter updates.
By Yu Cheng, Yongkang Hu, Jiuan Zhou, Yushuo Zhang, Yihang Chen, Huichi Zhou, Mingang Chen, Zhizhong Zhang, Kun Shao, Yuan Xie, Zhaoxia Yin
arXiv:2606. 22686v2 Announce Type: replace-cross Abstract: Modern Large Language Models (LLMs) rely on extensive safety alignment, yet the mechanistic basis of refusal remains opaque.
By Shivam Ratnakar, Kartikeya Vats
arXiv:2606. 05805v1 Announce Type: new Abstract: LLM-based guardrails typically safeguard agents by evaluating proposed actions or inputs before execution, producing safety signals such as binary allow/deny decisions, risk categories, and/or explanatory rationales about potential policy violations.
By Yuhao Sun, Jiacheng Zhang, Shaanan Cohney, Zhexin Zhang, Feng Liu, Xingliang Yuan
arXiv:2607. 16242v1 Announce Type: cross Abstract: Fine-Tuning-as-a-Service (FTaaS) platforms let users train large language models (LLMs) on customized tasks, but this pipeline could erode models' safety alignment.
By Changyue Li, Jiaming He, Youliang Yuan, Jialin Wu, Boxi Yu, Zhicong Huang, Pinjia He