返回全部动态
CRAFTER：黑盒预测器纠错特征发现代理
原标题：When Do Corrective Features Help? An Agent for Corrective Feature Discovery on Black-Box Forecasters
AI 摘要
arXiv 上发布了一篇关于黑盒预测器纠错特征发现的研究论文，提出了 CRAFTER 代理。CRAFTER 通过组合搜索和 LLM 生成候选特征，并用验证门控筛选，在六个数据集和六个冻结骨干网络上超越了专用特征工程系统，将最弱骨干的错误率降低最多 27%。该研究为模型纠错提供了新思路。
以上摘要由 AI 生成，可能存在误差。事实请以原文为准。
正文节选
Computer Science > Machine Learning Title:When Do Corrective Features Help? An Agent for Corrective Feature Discovery on Black-Box Forecasters View PDF HTML (experimental) Abstract:Frozen pretrained forecasters often fail in structured, recurring ways that are costly to repair through fine-tuning. We study corrective feature discovery: mining interpretable features of a frozen forecaster's residual to drive a lightweight post-hoc corrector. Prior automated feature engineering models
发布时间：2026-08-07 12:00
抓取时间：2026-08-07 14:19
来源机构：arXiv