arXiv 2609.39982 · cs.CL · 2026-09-30
Mid-Harness: Scaling Actions Between Model and Harness for Terminal Agents
Minki Kang, Ryo Hachiuma, Shaokun Zhang, Subhashree Radhakrishnan, Yonggan Fu, Jindong Jiang, Mingjie Liu, Ehsan Hosseini-Asl 等 11 人
NVIDIA
在最新榜 #119 本站收录于 2026-10-21
一句话要点
Mid-Harness 在模型与执行框架之间先采样并验证候选命令再执行，提高终端 Agent 的可靠性。
Mid-Harness samples and verifies candidate actions before execution, improving terminal-agent reliability.
阅读 arXiv 原文 PDF Hugging Face 讨论 项目页
主题与学科
同主题：智能体
- Raven: The Harness of Harnesses for Composable Agentic Intelligence ▲748
- False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents ▲642
- BizAgentBench: Unveiling Hidden Challenges in Deploying Long-Horizon Agents for Real-World Business ▲490
- Omni-IO Skills: Harnessing Your Agent Omni-Native ▲485
- Player-1: Learning to Play from Synthetic Experience ▲348
数据说明
元数据来自 arXiv API（已核实）；点赞来自 Hugging Face 每日论文；引用来自 Semantic Scholar。最近刷新 2026-10-22。
要点由本站根据摘要自写，不转载摘要原文，请阅读原文。