Evolving Agents in the Dark: Retrospective Harness Optimization via Self-Preference

• URL: http://arxiv.org/abs/2606.05922v2
• Date: Wed, 10 Jun 2026 05:04:53 GMT
• Title: Evolving Agents in the Dark: Retrospective Harness Optimization via Self-Preference
• Abstract summary: We introduce Retrospective Harness Optimization (RHO), a self-supervised method that optimize agent harness using only past trajectories.<n>RHO selects a diverse coreset of challenging tasks from past trajectories and re-solves them in parallel.<n>A single optimization round improves the pass rate on SWE-Bench Pro from 59% to 78% without any external grading.
• Score: 51.55077364626896
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• 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.

Related papers

• AutoSaddler: Automatic Harness Optimization with Durable Updates from Agent Execution Traces [61.51487329124135]
  AutoSaddler is a framework that formulates harness improvement as an offline learning problem.<n>It iteratively updates the harness using failure signals from mini-batches.<n>Experiments on GAIA2, SWE-Bench Pro, and Terminal-Bench 2.0 show that AutoSaddler substantially improves agent performance.
  arXiv  Detail & Related papers  (2026-08-24T09:45:08Z)
• HarnessOpt-Bench: Evaluating LLMs at Harness Optimization [3.846854817984248]
  We introduce a benchmark for end-to-end harness optimization under expensive evaluation regimes.<n>A trusted execution environment enforces the evaluation boundary, target-agent resource use, and preserves candidate versions for audit.<n>Experiment results show that models separate more than the coding harnesses they act through, native harnesses are not consistently superior, and gains vary substantially across tasks.
  arXiv  Detail & Related papers  (2026-08-06T17:21:05Z)
• Co-Harness: Co-Evolving Harnesses and Model Weights for LLM Agents [37.91666123749406]
  Co-Harness is a framework that jointly optimize the agent harness and model parameters during post-training.<n>An autonomous case study shows that Co-Harness can recover from system crashes, improve inference efficiency, and discover ensemble strategies without human intervention.
  arXiv  Detail & Related papers  (2026-07-17T02:39:57Z)
• Recursive Harness Self-Improvement [28.432258880153125]
  Under model--harness co-evolution, harnesses are data-generating components whose execution traces can shape future foundation models.<n>We introduce Recursive Harness Self-Improvement (RHI), which represents the harness as a prompt-level specification of the agent loop.<n>RHI iteratively refines it using pairwise feedback over its own revision history.
  arXiv  Detail & Related papers  (2026-07-17T00:21:19Z)
• Do Agent Optimizers Compound? A Continual-Learning Evaluation on Terminal-Bench 2.0 [53.71882250666667]
  Most reported gains from agent-optimization methods are one-shot.<n>This does not test the setting that matters for deployed agents.<n>We compare three approaches to agent-harness optimization.
  arXiv  Detail & Related papers  (2026-07-15T16:36:04Z)
• ATLAS : Adaptive Self-Evolutionary Research Agent with Task-Distributed Multi-LLM Supporters [6.13905106667213]
  ATLAS is a task-distributed framework that iteratively develops a lightweight research agent.<n>Our core algorithm, Evolving Direct Preference Optimization (EvoDPO), adaptively updates the phase-indexed reference policy.<n>Results show that ATLAS improves stability and performance over a static single-agent baseline.
  arXiv  Detail & Related papers  (2026-02-02T19:23:33Z)
• Guided by Trajectories: Repairing and Rewarding Tool-Use Trajectories for Tool-Integrated Reasoning [65.10602992874787]
  AutoTraj is a framework that automatically learns TIR by repairing and rewarding tool-use trajectories.<n> Experiments on real-world benchmarks demonstrate the effectiveness of AutoTraj.
  arXiv  Detail & Related papers  (2026-01-30T14:42:04Z)
• ET-Agent: Incentivizing Effective Tool-Integrated Reasoning Agent via Behavior Calibration [68.89572566071575]
  ETAgent is a training framework for calibrating agent's tool-use behavior.<n>It is designed to progressively calibrate erroneous behavioral patterns to optimal behaviors.
  arXiv  Detail & Related papers  (2026-01-11T11:05:26Z)
• ROAD: Reflective Optimization via Automated Debugging for Zero-Shot Agent Alignment [1.6968020497268546]
  ROAD is a novel framework that treats optimization as a dynamic debug investigation rather than a search.<n>Road is highly sample-efficient, achieving a 5.6 percent increase in success rate and a 3.8 percent increase in search accuracy.<n>These findings suggest that mimicking the human engineering loop of failure analysis and patching offers a viable, data-efficient alternative to resource-intensive training.
  arXiv  Detail & Related papers  (2025-12-30T07:31:34Z)
• Agent-R: Training Language Model Agents to Reflect via Iterative Self-Training [18.896813839389893]
  We propose an iterative self-training framework, Agent-R, that enables language Agent to Reflect on the fly.<n>Unlike traditional methods that reward or penalize actions based on correctness, Agent-R leverages MCTS to construct training data that recover correct trajectories from erroneous ones.<n>Our findings demonstrate that Agent-R continuously improves the model's ability to recover from errors and enables timely error correction.
  arXiv  Detail & Related papers  (2025-01-20T11:46:04Z)
• Self-Steering Optimization: Autonomous Preference Optimization for Large Language Models [79.84205827056907]
  We present Self-Steering Optimization ($SSO$), an algorithm that autonomously generates high-quality preference data.<n>$SSO$ employs a specialized optimization objective to build a data generator from the policy model itself, which is used to produce accurate and on-policy data.<n>Our evaluation shows that $SSO$ consistently outperforms baselines in human preference alignment and reward optimization.
  arXiv  Detail & Related papers  (2024-10-22T16:04:03Z)
• REBEL: Reward Regularization-Based Approach for Robotic Reinforcement Learning from Human Feedback [61.54791065013767]
  A misalignment between the reward function and human preferences can lead to catastrophic outcomes in the real world.<n>Recent methods aim to mitigate misalignment by learning reward functions from human preferences.<n>We propose a novel concept of reward regularization within the robotic RLHF framework.
  arXiv  Detail & Related papers  (2023-12-22T04:56:37Z)
• Learning to Optimize for Reinforcement Learning [58.01132862590378]
  Reinforcement learning (RL) is essentially different from supervised learning, and in practice, these learneds do not work well even in simple RL tasks. Agent-gradient distribution is non-independent and identically distributed, leading to inefficient meta-training. We show that, although only trained in toy tasks, our learned can generalize unseen complex tasks in Brax.
  arXiv  Detail & Related papers  (2023-02-03T00:11:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.