将知识图谱与细粒度向量结合，提升复杂查询的文档重排序效果。
QDER: Query-Specific Document and Entity Representations for Multi-Vector Document Re-Ranking
- 保留每个词和实体的独立表示，最后才做聚合，实现精准匹配。
- 在TREC Robust 2004上比最强基线提升36%的nDCG@20。
- 特别擅长处理困难查询，适合需要高精度检索的场景。
神经信息检索发展出两条路径：基于实体的方法利用知识图谱，多向量模型捕捉细粒度语义。本文提出QDER，一种统一这两种方法的神经重排序模型，通过将知识图谱语义融入多向量模型中。其核心创新在于对查询-文档关系的建模：不依赖聚合嵌入计算相似度，而是全程保持每个词和实体的独立表示，仅在最终打分阶段进行聚合——我们称之为“晚聚合”。首先通过学习的注意力模式转换这些细粒度表示，再应用精心设计的数学运算实现精确匹配。在五个标准基准测试上实验表明，QDER取得显著性能提升，在TREC Robust 2004上相比最强基线nDCG@20提升36%，其他数据集也有类似改善。尤其在困难查询上表现突出，nDCG@20达到0.70，而传统方法完全失败（nDCG@20 = 0.0），为未来实体感知检索研究奠定基础。
原文摘要 · Abstract (English)
Neural IR has advanced through two distinct paths: entity-oriented approaches leveraging knowledge graphs and multi-vector models capturing fine-grained semantics. We introduce QDER, a neural re-ranking model that unifies these approaches by integrating knowledge graph semantics into a multi-vector model. QDER's key innovation lies in its modeling of query-document relationships: rather than computing similarity scores on aggregated embeddings, we maintain individual token and entity representations throughout the ranking process, performing aggregation only at the final scoring stage - an approach we call "late aggregation." We first transform these fine-grained representations through learned attention patterns, then apply carefully chosen mathematical operations for precise matches. Experiments across five standard benchmarks show that QDER achieves significant performance gains, with improvements of 36% in nDCG@20 over the strongest baseline on TREC Robust 2004 and similar improvements on other datasets. QDER particularly excels on difficult queries, achieving an nDCG@20 of 0.70 where traditional approaches fail completely (nDCG@20 = 0.0), setting a foundation for future work in entity-aware 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。