用推理计划反馈优化检索，提升复杂问答的准确率
Adaptive Retrieval for Reasoning-Intensive Retrieval
- 将推理过程转化为检索反馈信号，动态调整文档召回
- 在复杂问答任务上比基线高5.6个百分点
- 适合需要深度推理的智能问答系统开发者
我们研究如何利用自适应检索确保在推理密集型检索中获取足够的‘桥梁文档’——这些文档虽不直接相关于初始查询，但对推理过程至关重要。现有基于推理的重排序流水线虽试图暴露这些文档，却存在召回受限的问题。简单地将自适应检索引入此类流水线常导致规划错误传播。为此，我们提出REPAIR框架，通过将推理计划重新用作密集反馈信号，实现自适应检索的中途修正，从而检索支持关键推理路径的文档。在推理密集型检索和复杂问答任务上的实验表明，该方法相比现有基线提升5.6个百分点。
原文摘要 · Abstract (English)
We study leveraging adaptive retrieval to ensure sufficient "bridge" documents are retrieved for reasoning-intensive retrieval. Bridge documents are those that contribute to the reasoning process yet are not directly relevant to the initial query. While existing reasoning-based reranker pipelines attempt to surface these documents in ranking, they suffer from bounded recall. Naive solution with adaptive retrieval into these pipelines often leads to planning error propagation. To address this, we propose REPAIR, a framework that bridges this gap by repurposing reasoning plans as dense feedback signals for adaptive retrieval. Our key distinction is enabling mid-course correction during reranking through selective adaptive retrieval, retrieving documents that support the pivotal plan. Experimental results on reasoning-intensive retrieval and complex QA tasks demonstrate that our method outperforms existing baselines by 5.6%pt.
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。