发表机构
Chung-Ang University（中央大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文针对多跳问答中现有提示压缩方法无法捕捉多跳推理步骤的问题，提出无训练框架IterCOMP，经实验验证其可提升问答指标并减少token预算。
AI 中文摘要
多跳问答需要跨多个证据片段进行复杂推理，这往往会使检索增强生成（RAG）系统因冗长且含噪声的上下文而不堪重负，进而损害效率与准确性。现有提示压缩方法虽试图解决该问题，但通常针对单轮查询设计，无法捕捉相互关联的推理步骤。本文提出IterCOMP，一种统一的无训练提示压缩框架，在迭代压缩循环中融入多跳推理。IterCOMP将文档分解为证据片段，评估问题可答性，并生成针对性的后续问题以迭代整合必要证据，生成紧凑、面向推理的提示。在MusiQue、2WikiMultiHopQA和HotpotQA上的实验表明，IterCOMP在精确匹配（Exact Match）和F1分数上实现显著提升，同时减少了token预算，优于现有基线，且在推理复杂度增加时表现出鲁棒性。
英文摘要
Multi-hop question answering requires complex reasoning across multiple evidence segments, which often overwhelms retrieval-augmented generation systems with lengthy and noisy contexts, thereby undermining both efficiency and accuracy. While existing prompt compression methods attempt to address this issue, they are typically designed for single-turn queries and fail to capture interdependent reasoning steps. We propose IterCOMP, a unified, training-free prompt compression framework that incorporates multi-hop reasoning within an iterative compression loop. IterCOMP decomposes documents into evidence segments, evaluates question answerability, and generates targeted follow-up questions to iteratively integrate essential evidence, producing a compact, reasoning-oriented prompt. Experiments on MusiQue, 2WikiMultiHopQA, and HotpotQA demonstrate that IterCOMP achieves substantial improvements in Exact Match and F1 scores while reducing the token budget, outperforming existing baselines and exhibiting robustness as reasoning complexity increases.
CommentsACL 2026 Main Conference
DOI：10.18653/v1/2026.acl-long.1559
