Episode
Evolving Agents in the Dark: Retrospective Harness Optimization via Self-Preference
- Podcast
- Daily Paper Cast
- Published
- Jun 11, 2026
- Duration seconds
- 1269
- Processing state
not_requested- Canonical source
- https://share.transistor.fm/s/c45684dd
Actions
POST https://stenobird.com/v1/public/podcasts/daily-paper-cast-7079649/episodes/evolving-agents-in-the-dark-retrospective-harness-optimization-via-self-preference/transcription-requests
Idempotently request low-priority transcript generation for this episode.GET https://stenobird.com/podcast/daily-paper-cast-7079649/evolving-agents-in-the-dark-retrospective-harness-optimization-via-self-preference.md
Read the agent-friendly Markdown representation of this episode resource.
Summary
🤗 Upvotes: 48 | cs.AI, cs.CL, cs.LG Authors: Wenbo Pan, Shujie Liu, Chin-Yew Lin, Jingying Zeng, Xianfeng Tang, Xiangyang Zhou, Yan Lu, Xiaohua Jia Title: Evolving Agents in the Dark: Retrospective Harness Optimization via Self-Preference Arxiv: http://arxiv.org/abs/2606.05922v2 Abstract: AI agents rely on a harness of skills, tools, and workflows to solve complex problems. Continually improving this harness is essential for adapting to new tasks. However, existing optimization methods typically require ground-truth validation sets, yet such labeled data is difficult to acquire in practical deployment settings. To address this problem, we introduce Retrospective Harness Optimization (RHO), a self-supervised method that optimizes the agent harness using only past trajectories. Specifically, RHO selects a diverse coreset of challenging tasks from past trajectories and re-solves them in parallel. The agent analyzes these rollouts using self-validation and self-consistency, then generates candidate harness updates and selects the most effective one by its own pairwise self-preference. We evaluate RHO across three diverse domains, spanning software engineering, technical work, and knowledge work. Notably, a single optimization round improves the pass rate on SWE-Bench Pro from 59% to 78% without any external grading. Furthermore, our analysis demonstrates that RHO effectively targets prior failure modes. As a result, the optimized harness alters the agent's behavior patterns and sustains higher accuracy during long-horizon sessions.