SafeEvolve is an experience-driven framework that co‑evolves a harness and policy to improve safety alignment for LLM‑based agents. It uses on‑policy trajectory safety evidence to update safety prompts and hierarchical skills, producing auditable harness artifacts. The policy is trained via a two‑stage SFT‑RL pipeline that bootstraps with the evolved harness and then refines behavior through verifier‑decomposed rewards, yielding a better safety‑utility tradeoff on benchmarks such as AgentDojo.
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:2609.01595v1 Announce Type: cross
Abstract: We develop a framework for mechanism design with AI agents whose alignment (preferences) and capabilities (feasible actions and information) are unkn...
By Dirk Bergemann, Andrew Koh, Stephen Morris
SEABench is a benchmark designed to study endogenous misalignment in self‑evolving large language model agents. It contains 48 longitudinal task sequences across various evolution surfaces, task domains, and harm types, and includes an adaptive trajectory discovery pipeline that probes for failures while preserving task intent. Evaluations show that self‑evolution improves task completion rates but often introduces safety failures absent in non‑evolving baselines, with divergent safety behaviors reflected in agents’ chain‑of‑thought reasoning that can be monitored to mitigate unsafe actions.
arXiv:2609.35596v2 Announce Type: replace-cross
Abstract: Self-evolving LLM agents have gained prominence for their ability to improve after deployment by modifying their harness, including their con...
By Saswat Das, Parvati Viswanathan, Daniel Donnelly, Chang Huang, Sahar Abdelnabi, Ferdinando Fioretto
arXiv:2607. 04613v1 Announce Type: new Abstract: Autonomous agents are moving from sandboxed text generators to operators of code, data, and physical infrastructure, and they increasingly learn while deployed.
By Xue Qin, Simin Luan, Cong Yang, Zhijun Li
arXiv:2606. 02380v1 Announce Type: cross Abstract: As LLM-based agents expand their operational scope, reliability becomes a prerequisite for real-world deployment.
By Yuyan Bu, Haowei Li, Qirui Zheng, Bowen Dong, Kaiyue Yang, Jiaming Ji, Yingshui Tan, Wenxin Li, Yaodong Yang, Juntao Dai
arXiv:2607. 18366v1 Announce Type: new Abstract: Large language models (LLMs) serving as planners in tool-using autonomous agents introduce dynamic reliability risks in multi-turn execution.
By Shasha Yu, Fiona Carroll, Barry L. Bentley
arXiv:2610.10285v1 Announce Type: cross
Abstract: In the wake of the increasingly public discussion about AI alignment, recent work has tried to propose specific AI architectures that behave safely....
By Leo Zeitler, Jack Richings, Victoria Nockles
arXiv:2605. 22748v2 Announce Type: replace-cross Abstract: Autonomous systems have achieved superhuman performance in isolation or simulation, yet they remain brittle in shared, dynamic real-world spaces.
By Ismail Geles, Leonard Bauersfeld, Markus Wulfmeier, Davide Scaramuzza
arXiv:2606. 07805v1 Announce Type: new Abstract: The rapid evolution of Large Language Models (LLMs) from passive assistants to autonomous, execution-capable agents has introduced critical operational risks.
By Yiyang Zhao, Zhuo Zhang, Qingxuan Le, Lizhen Qu, Zenglin Xu
arXiv:2606. 03237v1 Announce Type: new Abstract: AI's central challenge is shifting from capability to coexistence.
By Rakshit S Trivedi, Natasha Jaques, Logan Cross, Alexander Sasha Vezhnevets, Joel Z Leibo