研究人员开发了一个自动化研究系统，该系统使用专业代理来创建有效的AI训练配方。该系统作为一个封闭的经验循环运行，每次试验都包括一个假设、代码编辑和结果，反馈会塑造后续的建议。代理自主编写代码、提交实验，并根据崩溃或准确性未达标等结果来改进配方，从而在各种基准测试中显著提高模型性能和效率。 AI
影响 自动化优化训练配方的发现过程，可能加速AI开发并提高模型性能。
排序理由 该集群包含一篇学术论文，详细介绍了自动化AI研究的新方法。
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 →
arXiv:2605.05724v1 Announce Type: cross Abstract: We study auto research as a closed empirical loop driven by external measurement. Each submitted trial carries a hypothesis, an executable code edit, an evaluator-owned outcome, and feedback that shapes the next proposal. The outp…
We study auto research as a closed empirical loop driven by external measurement. Each submitted trial carries a hypothesis, an executable code edit, an evaluator-owned outcome, and feedback that shapes the next proposal. The output is not a generated paper or a single model chec…