When Agents Fail: Model or Harness–Who's to Blame?with Shao Zhang, Hangfan Zhang, Yuanyuan Shi, Hao Li +71 versionW_bjdkr6t8v1 · current
Harness-of-Harness: Multi-Day Autonomous Software Development with Continual ImprovementHaoyang Yan, Min-le Su, Hangfan Zhang et al.0 citations2609.01481v12026arXiv