cs.AI · 2607.03935 Copy arXiv ID · Jul 4, 2026 Save Harness-Aware Self-Evolving: Co-Evolving Model Weights, Harness, and Task Solutions Authors: Haochen Luo , Yi Huang , Sichun Luo , Fengyuan Liu , Lei Li , Zefa Hu , Junlan Feng , Qi Liu
Organizations: School of Computing and Data Science, The University of Hong Kong · Grace Investment Machine · Jiutian Research, China Mobile
Abstract Self-evolving frameworks usually optimize task solutions while treating the surrounding harness as fixed. We introduce Harness-Aware Self-Evolving (HASE), an agentic reinforcement-learning framework in which a single model can generate task solutions or edit selected harness components in a multi-turn action space. HASE enables a single Qwen3-8B model to match the text-classification performance of a GPT-OSS-120B model that uses Claude Code as the harness proposer. In alpha factor mining, HASE outperforms the reported GPT-OSS-120B baseline. HASE also repairs imperfect evaluation components and converges to state-of-the-art performance in circle-packing algorithm discovery. These results show that HASE improves the harness and the solution through one unified agentic process.
Explore similar work May 28, 2026 · Minhua Lin, Juncheng Wu, Zijun Wang +14 Agent Harness Evolution
Aug 3, 2026 · Luan Zhang, Ruochen Zhou, Dandan Song +9 Agent Harness Evolution
Jun 12, 2026 · Tingyang Chen, Shuo Lu, Kang Zhao +11 Agent Harness Multi-Agent Evolution
May 28, 2026 · cs.AI J/K move · Enter open · S save
Minhua Lin, Juncheng Wu, Zijun Wang, Zhan Shi +13
1The Pennsylvania State University · 2UC Santa Cruz · 3Amazon · 4Emory University · 5UIUC
LLM agents are increasingly deployed as systems built around editable external harnesses, including prompts, skills, memories and tools, that shape task execution without changing model parameters. Harness self-evolution adapts such agents by updating these harnesses from execution evidence. Yet it remains unclear whether a model's base capability in task-solving predicts its capabilities in harness self-evolution: which models produce useful harness updates, and which actually benefit from them? We analyze two harness self-evolution capabilities: (i) harness-updating, the capability to produce useful persistent harness updates from execution evidence; (ii) harness-benefit, the capability to benefit from updated harnesses during task solving. Our analysis reveals two findings. First, harness-updating is flat in base capability: models from different capability tiers produce harness updates that lead to surprisingly similar gains; even Qwen3.5-9B's updates yield gains comparable to those of Claude Opus~4.6. Second, harness-benefit is non-monotonic in base capability: weak-tier models benefit little from updated harnesses, mid-tier models benefit most, and strong-tier models benefit less than mid-tier. We trace low gains at the weak tier to two failure modes: weak-tier models may fail to activate relevant harness artifacts, or activate them but fail to follow them faithfully. These findings suggest investing capability budget in the task-solving agent rather than the evolver, and targeting harness invocation and long-horizon instruction following in agent training. Our source code is publicly available at https://github.com/A-EVO-Lab/a-evolve/tree/release/harness-evolution.