Efficient Document Ranking With Learnable Late Interactions

Abstract

Cross-Encoder (CE) and Dual-Encoder (DE) models are two fundamental approaches for query-document relevance in information retrieval. To predict relevance, CE models use joint query-document embeddings, while DE models maintain factorized query and document embeddings; usually, the former has higher quality while the latter benefits from lower latency. Recently, late-interaction models have been proposed to realize more favorable latency-quality tradeoffs, by using a DE structure followed by a lightweight scorer based on query and document token embeddings. However, these lightweight scorers are often hand-crafted, and there is no understanding of their approximation power; further, such scorers require access to individual document token embeddings, which imposes an increased latency and storage burden. In this paper, we propose novel learnable late-interaction models (LITE) that resolve these issues. Theoretically, we prove that LITE is a universal approximator of continuous scoring

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyji2024efficient

Related papers

• MICE: Minimal Interaction Cross-encoders For Efficient Re-ranking (2026)0.00
• CODER: An Efficient Framework For Improving Retrieval Through Contextual Document Embedding Reranking (2021)7.16
• Learning Diverse Document Representations With Deep Query Interactions For Dense Retrieval (2022)2.51
• Can Cross Encoders Produce Useful Sentence Embeddings? (2025)0.00
• Adaptive Retrieval And Scalable Indexing For K-nn Search With Cross-encoders (2024)0.00
• Llm-augmented Retrieval: Enhancing Retrieval Models Through Language Models And Doc-level Embedding (2024)0.00
• Enhancing Documents With Multidimensional Relevance Statements In Cross-encoder Re-ranking (2023)0.00
• Query Encoder Distillation Via Embedding Alignment Is A Strong Baseline Method To Boost Dense Retriever Online Efficiency (2023)0.00