arXiv AI October 31, 2026 By Hau-Sing So, Daixian Liu, Weicai Long

Masked Erosion: How Acceptance Gates Lose Redundant Safeguards in Self-Improving Agents

The paper investigates how self‑improving large language model agents can inadvertently remove redundant safeguards through acceptance gates that evaluate edits based on task success alone. By systematically deleting safeguards in two agent environments and running 411 improvement loops under various gate rules, the authors show that when one safeguard is removed, others still block harm, yet the gate may still delete it if it improves task performance. Larger evaluation batches and certain optimization strategies (e.g., GEPA) can lead to the loss of multiple safeguards, exposing the system to harm when the harness is reused with a weaker model.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv AI

Sep 23

Self-Healing Harness for Runtime Oversight of Agent Self-Modification

The paper introduces a self‑healing harness that enforces admission control over language‑model agents’ self‑modifications. The harness runs a Detect‑Notice‑Heal‑Validate loop, allowing agents to propose rule changes that are only granted persistent authority after demonstrating improvement on a failure case without regressing on protected cases. Across 16 benchmark runs, the harness rejected many locally beneficial proposals that caused collateral regressions, while improving task‑completion scores and reliability.

By Sina Tayebati, Divake Kumar, Nastaran Darabi, Ranganath Krishnan, Amit Ranjan Trivedi

llms agents

arXiv AI

Jul 8

Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade

arXiv:2607. 06503v1 Announce Type: new Abstract: Large language model (LLM) agents solving multi-step tasks frequently commit to trajectories that are doomed to fail, yet continue to consume substantial inference compute before the failure becomes observable.

By Kai Ruan, Zihe Huang, Ziqi Zhou, Qianshan Wei, Xuan Wang, Hao Sun

llms agents

Hugging Face Trending Papers

Jul 7

Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade

Large language model (LLM) agents solving multi-step tasks frequently commit to trajectories that are doomed to fail, yet continue to consume substantial inference compute before the failure becomes observable. We show that failure is predictable early from the agent's internal representations: lightweight per-round probes on hidden activations anticipate eventual episode failure as early as the first interaction round, where scorers reading only the agent's observable behavior are barely better than chance.

llms agents

arXiv AI

Aug 18

No Task Fails Every Time: Why One-Shot Audits Are Structurally Blind to Agent Damage

arXiv:2608. 15286v1 Announce Type: cross Abstract: We introduce AgentRelBench, an environment-agnostic reliability instrument that computes ground-truth, severity-priced damage from database state diffs across repeated runs, with no LLM in the measurement path, demonstrated on EnterpriseOps-Gym.

By Shiven Khurdi

llms agents

arXiv AI

Aug 14

Practice Makes Unsafe: Skill Misevolution in Self-Improving LLM Agents

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

llms agents multimodal benchmarks safety

arXiv AI

Aug 7

When Self-Evolution Backfires: Pre-Commit Gating against Skill Contamination in LLM Agents

arXiv:2608. 05810v1 Announce Type: new Abstract: Self-evolving agents accumulate capability by distilling reusable skills from their execution trajectories, but we find this process is not monotonic: past a critical pool size, newly added skills degrade performance instead of improving it.

By Linfang Shang, Ming Xu, Yiding Sun, Tianle Xia, Lingxiang Hu, Lan Xu, Ning Zheng

llms agents benchmarks

One email a morning, machine-written

Related stories

Self-Healing Harness for Runtime Oversight of Agent Self-Modification

Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade

Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade

No Task Fails Every Time: Why One-Shot Audits Are Structurally Blind to Agent Damage

Practice Makes Unsafe: Skill Misevolution in Self-Improving LLM Agents

When Self-Evolution Backfires: Pre-Commit Gating against Skill Contamination in LLM Agents