Early Stage Sparse Retrieval With Entity Linking

Abstract

Despite the advantages of their low-resource settings, traditional sparse retrievers depend on exact matching approaches between high-dimensional bag-of-words (BoW) representations of both the queries and the collection. As a result, retrieval performance is restricted by semantic discrepancies and vocabulary gaps. On the other hand, transformer-based dense retrievers introduce significant improvements in information retrieval tasks by exploiting low-dimensional contextualized representations of the corpus. While dense retrievers are known for their relative effectiveness, they suffer from lower efficiency and lack of generalization issues, when compared to sparse retrievers. For a lightweight retrieval task, high computational resources and time consumption are major barriers encouraging the renunciation of dense models despite potential gains. In this work, we propose boosting the performance of sparse retrievers by expanding both the queries and the documents with linked entities in

Stats

• citations7
• S2 citations—
• github stars0
• HF likes0
• heat score6.77
• arxiv keyshehata2022early

Related papers

• Predicting Efficiency/effectiveness Trade-offs For Dense Vs. Sparse Retrieval Strategy Selection (2021)11.29
• Towards Competitive Search Relevance For Inference-free Learned Sparse Retrievers (2024)0.00
• Dyvo: Dynamic Vocabularies For Learned Sparse Retrieval With Entities (2024)5.84
• Faster Learned Sparse Retrieval With Guided Traversal (2022)11.29
• Improved Learned Sparse Retrieval With Corpus-specific Vocabularies (2024)6.34
• Slimrag: Retrieval Without Graphs Via Entity-aware Context Selection (2025)1.91
• Llm-augmented Retrieval: Enhancing Retrieval Models Through Language Models And Doc-level Embedding (2024)0.00
• End-to-end Retrieval With Learned Dense And Sparse Representations Using Lucene (2023)0.00