QDER: Query-specific Document And Entity Representations For Multi-vector Document Re-ranking

Abstract

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 improvem

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keychatterjee2025qder

Related papers

• SDR: Efficient Neural Re-ranking Using Succinct Document Representation (2021)3.58
• Incorporating Query Term Independence Assumption For Efficient Retrieval And Ranking Using Deep Neural Networks (2019)0.00
• Attention Grounded Enhancement For Visual Document Retrieval (2025)0.00
• Improving Bert-based Query-by-document Retrieval With Multi-task Optimization (2022)9.92
• SEINE: Segment-based Indexing For Neural Information Retrieval (2023)0.00
• Neural Ranking Models For Document Retrieval (2021)11.08
• CODER: An Efficient Framework For Improving Retrieval Through Contextual Document Embedding Reranking (2021)7.16
• MRNN: A Multi-resolution Neural Network With Duplex Attention For Document Retrieval In The Context Of Question Answering (2019)0.00