# A Dense Representation Framework for Lexical and Semantic Matching

arXiv: https://arxiv.org/abs/2206.09912 · HTML: https://mlsanity.com/paper/2206.09912/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Jun 2022.

## Built on by (most cited first, 8 of 8)

- [TRAQ: Trustworthy Retrieval Augmented Question Answering via Conformal Prediction](https://mlsanity.com/paper/2307.04642/) (2023)
- [CITADEL: Conditional Token Interaction via Dynamic Lexical Routing for Efficient and Effective Multi-Vector Retrieval](https://mlsanity.com/paper/2211.10411/) (2022)
- [Multimodal Learned Sparse Retrieval with Probabilistic Expansion Control](https://mlsanity.com/paper/2402.17535/) (2024)
- [Multi-Field Adaptive Retrieval](https://mlsanity.com/paper/2410.20056/) (2024)
- [Semi-Parametric Retrieval via Binary Bag-of-Tokens Index](https://mlsanity.com/paper/2405.01924/) (2024)
- [Rationale-Augmented Retrieval with Constrained LLM Re-Ranking for Task Discovery](https://mlsanity.com/paper/2510.05131/) (2025)
- [Gradient-free Task-Conditioned Retrieval for On-Device In-Context Learning](https://mlsanity.com/paper/2607.27766/) (2026)
- [Exploring In-Context Learning for Handwritten Text Recognition](https://mlsanity.com/paper/2609.37195/) (2026)

## Abstract (arXiv)

Lexical and semantic matching capture different successful approaches to text retrieval and the fusion of their results has proven to be more effective and robust than either alone. Prior work performs hybrid retrieval by conducting lexical and semantic matching using different systems (e.g., Lucene and Faiss, respectively) and then fusing their model outputs. In contrast, our work integrates lexical representations with dense semantic representations by densifying high-dimensional lexical representations into what we call low-dimensional dense lexical representations (DLRs). Our experiments show that DLRs can effectively approximate the original lexical representations, preserving effectiveness while improving query latency. Furthermore, we can combine dense lexical and semantic representations to generate dense hybrid representations (DHRs) that are more flexible and yield faster retrieval compared to existing hybrid techniques. In addition, we explore it jointly training lexical and semantic representations in a single model and empirically show that the resulting DHRs are able to combine the advantages of the individual components. Our best DHR model is competitive with state-of-the-art single-vector and multi-vector dense retrievers in both in-domain and zero-shot evaluation settings. Furthermore, our model is both faster and requires smaller indexes, making our dense representation framework an attractive approach to text retrieval. Our code is available at https://github.com/castorini/dhr.