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...
The article surveys safety concerns for self‑evolving agents that continually update their internal state, such as model parameters and memories, from new interactions. It introduces the SAVER framework, which tracks reusable influence, adaptation, violations, exposure, and response to assess whether safety properties persist as agents evolve. The survey finds that legitimate state can become unsafe when its persistence, authority, or scope expands beyond its original conditions, and highlights gaps in current research on descendant repair and longitudinal evaluation.
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.
arXiv:2606. 06114v1 Announce Type: new Abstract: Self-evolving agents improve through continual self-play and self-generated learning signals, but autonomous evolution can also cause capability degradation and safety drift.
arXiv:2608. 12851v1 Announce Type: new Abstract: Self-improving LLM agents convert successful trajectories into persistent cross-task state.
The paper introduces ANCHOR, an external supervisory framework driven by large language models (LLMs) that provides evaluative feedback at multiple stages of self‑evolving agents. By integrating ANCHOR into two open‑source self‑evolving agent frameworks, the authors demonstrate that it significantly improves safety performance while preserving core capabilities across coding, mathematical reasoning, and safety tasks. The study also finds that supervision based on execution results is especially effective and that increasing supervision frequency yields diminishing returns, offering practical guidance for future research.
Self-evolving agents convert interaction feedback into persistent artifacts, such as memories or skills, which in turn guide subsequent decisions. As these artifacts are iteratively updated throughout...
The paper introduces the concept of Evolutionary Safety for recursive self-improving AI, focusing on how safety properties evolve as an AI system and its successors change. It identifies key risks such as intent drift, error accumulation, and safety-property erosion, and presents a taxonomy covering agent state, model state, evaluation, environment, and update mechanisms. The authors propose methods for discovering and evaluating evolutionary risks, and outline governance principles for modification, selection, authorization, provenance, and recovery, while highlighting open problems for maintaining safety in persistent, adaptive, and recursively self-improving systems.
arXiv:2609.24663v1 Announce Type: new Abstract: Self-evolving agents convert interaction feedback into persistent artifacts, such as memories or skills, which in turn guide subsequent decisions. As t...
arXiv:2607. 26820v1 Announce Type: new Abstract: As large language models (LLMs) evolve from standalone assistants into autonomous agents, ensuring their safety requires shifting beyond pointwise risk assessment to understand how risks emerge and unfold over long-horizon trajectories.
HarnessEvolve is a self‑evolving framework that improves agent harnesses—prompts, skills, tools, and execution logic—by learning from reference trajectories. It separates execution, evaluation, optimization, and gating into independent modules, addressing credit assignment failure, shortcut learning, and catastrophic forgetting. The approach uses reference trajectories to extract error signals, applies quality and performance gates to candidate updates, and validates updates on held‑out data, consistently outperforming state‑of‑the‑art baselines across diverse benchmarks.
arXiv:2608.30429v1 Announce Type: new Abstract: LLM-based agent systems increasingly adopt skill-based architectures to reduce repetitive reasoning costs and improve stable, efficient task execution....