让机器能理解的模型，比人类用的更高效。
Agentic-imodels: Evolving agentic interpretability tools via autoresearch
- 用AI自研可被代理理解的模型，而非仅适合人看。
- 新模型在预测性能和代理可读性上双双提升73%。
- 适合构建自动数据分析系统的研究者使用。
自主数据科学（ADS）系统正快速提升其自主分析、拟合和解释数据的能力，未来可能由智能体承担大部分数据科学工作。然而，现有系统仍依赖为人设计的可解释性工具，而非为智能体优化。为此，我们提出Agentic-imodels——一个基于自研循环的智能体式研究框架，旨在演化出专为智能体可解释性设计的数据科学工具。具体而言，该框架开发了一套兼容scikit-learn的表格数据回归器，兼顾预测性能与一种基于大语言模型（LLM）的新可解释性度量。该度量通过一系列由LLM评分的测试，评估拟合模型的字符串表示是否“可模拟”：即仅凭阅读模型输出字符串，能否让LLM回答关于模型行为的问题。实验表明，所进化出的模型在预测性能与代理可解释性上均实现提升，并能在新数据集和新测试中泛化。此外，这些模型显著提升了端到端的自主数据科学系统表现，在BLADE基准上使Copilot CLI、Claude Code和Codex的性能最高提升73%。
原文摘要 · Abstract (English)
Agentic data science (ADS) systems are rapidly improving their capability to autonomously analyze, fit, and interpret data, potentially moving towards a future where agents conduct the vast majority of data-science work. However, current ADS systems use statistical tools designed to be interpretable by humans, rather than interpretable by agents. To address this, we introduce Agentic-imodels, an agentic autoresearch loop that evolves data-science tools designed to be interpretable by agents. Specifically, it develops a library of scikit-learn-compatible regressors for tabular data that are optimized for both predictive performance and a novel LLM-based interpretability metric. The metric measures a suite of LLM-graded tests that probe whether a fitted model's string representation is "simulatable" by an LLM, i.e. whether the LLM can answer questions about the model's behavior by reading its string output alone. We find that the evolved models jointly improve predictive performance and agent-facing interpretability, generalizing to new datasets and new interpretability tests. Furthermore, these evolved models improve downstream end-to-end ADS, increasing performance for Copilot CLI, Claude Code, and Codex on the BLADE benchmark by up to 73%
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。