通过分块聚合证据提升长文档检索效果
Lost in a Single Vector: Improving Long-Document Retrieval with Chunk Evidence Aggregation
- 将长文档切块编码后聚合，保留关键片段信息
- 在超4000词片段上，准确率从30%提升至90%
- 无需训练，适合现有系统直接部署
密集检索将一个查询向量与一个文档向量进行匹配。在长文档场景下，当一段短但关键的文本在文档编码过程中被弱化时，该方法可能失效。我们将其归因于文档端的早期压缩，并提出证据稀释指数（EDI）来衡量文档级表示相对于同一黄金文档内最强片段级证据的偏离程度。基于此，我们提出DICE（Document Inference via Chunk Evidence），一种无需训练的文档端策略：将文档切分为块，用冻结模型独立编码，再聚合回单个向量，同时保持标准的一对一接口。在LongEmbed数据集上，DICE在四种骨干模型上均提升检索性能，尤其在超过4000词的片段上增益显著：Dream, Passkey >4k从30.0提升至90.0，Needle >4k从23.3提升至74.0。在12,779个过滤样本中，DICE的EDI低于单向量基线的比例达92.8%。结果表明，文档级编码是长文档检索中实用且未被充分探索的优化杠杆。
原文摘要 · Abstract (English)
Dense retrieval ranks one query vector against one document vector. On long documents, this interface can fail when a short but decisive span is weakened during document encoding before ranking. We study this failure mode as document-side early compression and introduce the Evidence Dilution Index (EDI) to measure how far a document-level representation falls below the strongest chunk-level evidence within the same gold document. Guided by this view, we propose DICE (Document Inference via Chunk Evidence), a training-free document-side strategy that splits documents into chunks, encodes them independently with a frozen model, and aggregates them back into a single vector while preserving the standard one-query-one-document interface. On LongEmbed, DICE improves retrieval across four backbones, with the largest gains on slices beyond 4k tokens: for Dream, Passkey >4k rises from 30.0 to 90.0 and Needle >4k from 23.3 to 74.0. Across 12,779 filtered samples, DICE yields lower EDI than the single-vector baseline in 92.8% of cases. These results establish document-level encoding as a practical and underexplored lever for long-document retrieval.
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。