PulseAugur
中
实时 11:55:32
English(EN) Harness-of-Harness: Multi-Day Autonomous Software Development with Continual Improvement

新框架赋能 LLM 代理自主开发和改进软件

作者 PulseAugur 编辑部 · [2 个来源] ·

研究人员推出 Harness-of-Harness (HoH) 框架，旨在通过使基于 LLM 的编码代理能够持续改进软件来增强自主软件开发能力。HoH 将开发过程结构化为迭代的规划-编码-测试循环，平衡修复与能力增长，并将任务分解为小型、可验证的增量。在 GameCraft-Bench、FrontierSWE 和 ProgramBench 等基准测试中，HoH 表现出显著的性能提升，平均比独立 harness 提高了 52.25%。在为期多天的部署中，HoH 自主开发了一个完整的笫一人称射击游戏。 AI

影响 该框架可以通过 LLM 代理显著加速复杂软件系统的开发，减少对人工干预的需求。

排序理由 该集群包含一篇详细介绍自主软件开发新框架的研究论文。

在 Hugging Face Daily Papers 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

新框架赋能 LLM 代理自主开发和改进软件

本文如何被排名

Signal score
2 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
该集群包含一篇详细介绍自主软件开发新框架的研究论文。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
1 days old
Coverage has settled into its steady-state source set.

完整方法见我们的编辑标准。

报道来源 [2]

  1. arXiv cs.AI TIER_1 English(EN) · Haoyang Yan, Min-le Su, Hangfan Zhang, Zhanhao Li, Chen Zhang, Shao Zhang, Yang Chen, Lei Bai, Shuyue Hu ·

    Harness-of-Harness：多日自主软件开发与持续改进

    arXiv:2609.01481v1 Announce Type: new Abstract: This paper studies autonomous software development, in which LLM-based coding agents transform high-level requirements into complete, functional, and usable software systems without human intervention. We introduce Harness-of-Harnes…

  2. Hugging Face Daily Papers TIER_1 English(EN) ·

    Harness-of-Harness：多日自主软件开发与持续改进

    Harness-of-Harness enables autonomous coding agents to iteratively improve software through structured planning-coding-testing loops, yielding substantial performance gains and fully developed applications.

相关实体

相关话题