# Generative Retrieval as Dense Retrieval

arXiv: https://arxiv.org/abs/2306.11397 · HTML: https://mlsanity.com/paper/2306.11397/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Jun 2023.

## Built on by (most cited first, 6 of 6)

- [From Matching to Generation: A Survey on Generative Information Retrieval](https://mlsanity.com/paper/2404.14851/) (2024)
- [GENIUS: A Generative Framework for Universal Multimodal Search](https://mlsanity.com/paper/2503.19868/) (2025)
- [ParamMem: Augmenting Language Agents with Parametric Reflective Memory](https://mlsanity.com/paper/2602.23320/) (2026)
- [Bottleneck-Minimal Indexing for Generative Document Retrieval](https://mlsanity.com/paper/2405.10974/) (2024)
- [Invar-RAG: Invariant LLM-aligned Retrieval for Better Generation](https://mlsanity.com/paper/2411.07021/) (2024)
- [Foundations of GenIR](https://mlsanity.com/paper/2501.02842/) (2025)

## Abstract (arXiv)

Generative retrieval is a promising new neural retrieval paradigm that aims to optimize the retrieval pipeline by performing both indexing and retrieval with a single transformer model. However, this new paradigm faces challenges with updating the index and scaling to large collections. In this paper, we analyze two prominent variants of generative retrieval and show that they can be conceptually viewed as bi-encoders for dense retrieval. Specifically, we analytically demonstrate that the generative retrieval process can be decomposed into dot products between query and document vectors, similar to dense retrieval. This analysis leads us to propose a new variant of generative retrieval, called Tied-Atomic, which addresses the updating and scaling issues by incorporating techniques from dense retrieval. In experiments on two datasets, NQ320k and the full MSMARCO, we confirm that this approach does not reduce retrieval effectiveness while enabling the model to scale to large collections.