ReFeed：基于反馈反思推理的多维度摘要精炼

ReFeed: Multi-dimensional Summarization Refinement with Reflective Reasoning on Feedback

March 27, 2025

作者: Taewon Yun, Jihwan Oh, Hyangsuk Min, Yuho Lee, Jihwan Bang, Jason Cai, Hwanjun Song

cs.AI

摘要

在多维度扩展时，摘要精炼面临诸多挑战。本文提出ReFeed，一种强大的摘要精炼流程，通过反馈的反思性推理来增强多个维度。为此，我们发布了SumFeed-CoT，一个大规模基于长链推理（Long-CoT）的数据集，专为训练具备反思推理能力的轻量级模型而优化。实验揭示了维度数量、反馈暴露程度及推理策略如何影响精炼效果，强调反思性推理与同时处理多重反馈对于缓解维度间权衡至关重要。此外，ReFeed对噪声反馈及反馈顺序表现出良好的鲁棒性。最后，我们的发现强调，以恰当目标和指导原则创建数据是构建有效推理的基石。数据集与模型将予以公开。

English

Summarization refinement faces challenges when extending to multi-dimension. In this paper, we introduce ReFeed, a powerful summarization refinement pipeline that enhances multiple dimensions through reflective reasoning on feedback. To achieve this, we release SumFeed-CoT, a large-scale Long-CoT-based dataset optimized for training a lightweight model with reflective reasoning. Our experiments reveal how the number of dimensions, feedback exposure, and reasoning policy influence refinement performance, highlighting reflective reasoning and simultaneously addressing multiple feedback is crucial to mitigate trade-off between dimensions. Furthermore, ReFeed is robust to noisy feedback and feedback order. Lastly, our finding emphasizes that creating data with a proper goal and guideline constitutes a fundamental pillar of effective reasoning. The dataset and model will be released.