# A Replication Study of Dense Passage Retriever

arXiv: https://arxiv.org/abs/2104.05740 · HTML: https://mlsanity.com/paper/2104.05740/

## Key facts

- Cited by 37 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 14 references that are in the map.
- First posted on arXiv 12 Apr 2021.

## Builds on (most cited first, 10 of 10)

- [Dense Passage Retrieval for Open-Domain Question Answering](https://mlsanity.com/paper/2004.04906/) (2020)
- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval](https://mlsanity.com/paper/2007.00808/) (2020)
- [Reading Wikipedia to Answer Open-Domain Questions](https://mlsanity.com/paper/1704.00051/) (2017)
- [Generation-Augmented Retrieval for Open-domain Question Answering](https://mlsanity.com/paper/2009.08553/) (2020)
- [End-to-End Open-Domain Question Answering with BERTserini](https://mlsanity.com/paper/1902.01718/) (2019)
- [Improving Efficient Neural Ranking Models with Cross-Architecture Knowledge Distillation](https://mlsanity.com/paper/2010.02666/) (2020)
- [Pyserini: An Easy-to-Use Python Toolkit to Support Replicable IR Research with Sparse and Dense Representations](https://mlsanity.com/paper/2102.10073/) (2021)
- [Distilling Dense Representations for Ranking using Tightly-Coupled Teachers](https://mlsanity.com/paper/2010.11386/) (2020)
- [RepBERT: Contextualized Text Embeddings for First-Stage Retrieval](https://mlsanity.com/paper/2006.15498/) (2020)

## Built on by (most cited first, 30 of 37)

- [Unsupervised Dense Information Retrieval with Contrastive Learning](https://mlsanity.com/paper/2112.09118/) (2021)
- [Improving Passage Retrieval with Zero-Shot Question Generation](https://mlsanity.com/paper/2204.07496/) (2022)
- [Mr. TyDi: A Multi-lingual Benchmark for Dense Retrieval](https://mlsanity.com/paper/2108.08787/) (2021)
- [ReACC: A Retrieval-Augmented Code Completion Framework](https://mlsanity.com/paper/2203.07722/) (2022)
- [How to Train Your DRAGON: Diverse Augmentation Towards Generalizable Dense Retrieval](https://mlsanity.com/paper/2302.07452/) (2023)
- [A Few Brief Notes on DeepImpact, COIL, and a Conceptual Framework for Information Retrieval Techniques](https://mlsanity.com/paper/2106.14807/) (2021)
- [Salient Phrase Aware Dense Retrieval: Can a Dense Retriever Imitate a Sparse One?](https://mlsanity.com/paper/2110.06918/) (2021)
- [Tevatron: An Efficient and Flexible Toolkit for Dense Retrieval](https://mlsanity.com/paper/2203.05765/) (2022)
- [Learning to Retrieve Passages without Supervision](https://mlsanity.com/paper/2112.07708/) (2021)
- [s3: You Don't Need That Much Data to Train a Search Agent via RL](https://mlsanity.com/paper/2505.14146/) (2025)
- [Graph Reasoning for Question Answering with Triplet Retrieval](https://mlsanity.com/paper/2305.18742/) (2023)
- [Challenges in Generalization in Open Domain Question Answering](https://mlsanity.com/paper/2109.01156/) (2021)
- [A Proposed Conceptual Framework for a Representational Approach to Information Retrieval](https://mlsanity.com/paper/2110.01529/) (2021)
- [TextGenSHAP: Scalable Post-hoc Explanations in Text Generation with Long Documents](https://mlsanity.com/paper/2312.01279/) (2023)
- [textttMixGR: Enhancing Retriever Generalization for Scientific Domain through Complementary Granularity](https://mlsanity.com/paper/2407.10691/) (2024)
- [A Study on the Efficiency and Generalization of Light Hybrid Retrievers](https://mlsanity.com/paper/2210.01371/) (2022)
- [AlignCoder: Aligning Retrieval with Target Intent for Repository-Level Code Completion](https://mlsanity.com/paper/2601.19697/) (2026)
- [Automated MeSH Term Suggestion for Effective Query Formulation in Systematic Reviews Literature Search](https://mlsanity.com/paper/2209.08687/) (2022)
- [Optimizing High-Level Synthesis Designs with Retrieval-Augmented Large Language Models](https://mlsanity.com/paper/2410.07356/) (2024)
- [Q-PEFT: Query-dependent Parameter Efficient Fine-tuning for Text Reranking with Large Language Models](https://mlsanity.com/paper/2404.04522/) (2024)
- [Task-Aware Specialization for Efficient and Robust Dense Retrieval for Open-Domain Question Answering](https://mlsanity.com/paper/2210.05156/) (2022)
- [Rethinking On-policy Optimization for Query Augmentation](https://mlsanity.com/paper/2510.17139/) (2025)
- [Hybrid Inverted Index Is a Robust Accelerator for Dense Retrieval](https://mlsanity.com/paper/2210.05521/) (2022)
- [Citation-Grounded Code Comprehension: Preventing LLM Hallucination Through Hybrid Retrieval and Graph-Augmented Context](https://mlsanity.com/paper/2512.12117/) (2025)
- [Retrieval Augmentation for T5 Re-ranker using External Sources](https://mlsanity.com/paper/2210.05145/) (2022)
- [Building Interpretable and Reliable Open Information Retriever for New Domains Overnight](https://mlsanity.com/paper/2308.04756/) (2023)
- [M3: A Multi-Task Mixed-Objective Learning Framework for Open-Domain Multi-Hop Dense Sentence Retrieval](https://mlsanity.com/paper/2403.14074/) (2024)
- [Know When to Fuse: Investigating Non-English Hybrid Retrieval in the Legal Domain](https://mlsanity.com/paper/2409.01357/) (2024)
- [Cognis: Context-Aware Memory for Conversational AI Agents](https://mlsanity.com/paper/2604.19771/) (2026)
- [NuggetIndex: Governed Atomic Retrieval for Maintainable RAG](https://mlsanity.com/paper/2604.27306/) (2026)

## Abstract (arXiv)

Text retrieval using learned dense representations has recently emerged as a promising alternative to "traditional" text retrieval using sparse bag-of-words representations. One recent work that has garnered much attention is the dense passage retriever (DPR) technique proposed by Karpukhin et al. (2020) for end-to-end open-domain question answering. We present a replication study of this work, starting with model checkpoints provided by the authors, but otherwise from an independent implementation in our group's Pyserini IR toolkit and PyGaggle neural text ranking library. Although our experimental results largely verify the claims of the original paper, we arrived at two important additional findings that contribute to a better understanding of DPR: First, it appears that the original authors under-report the effectiveness of the BM25 baseline and hence also dense--sparse hybrid retrieval results. Second, by incorporating evidence from the retriever and an improved answer span scoring technique, we are able to improve end-to-end question answering effectiveness using exactly the same models as in the original work.