# Unsupervised Document Expansion for Information Retrieval with Stochastic Text Generation

arXiv: https://arxiv.org/abs/2105.00666 · HTML: https://mlsanity.com/paper/2105.00666/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 May 2021.

## Built on by (most cited first, 9 of 9)

- [Augmenting Document Representations for Dense Retrieval with Interpolation and Perturbation](https://mlsanity.com/paper/2203.07735/) (2022)
- [Autoregressive Sign Language Production: A Gloss-Free Approach with Discrete Representations](https://mlsanity.com/paper/2309.12179/) (2023)
- [PODTILE: Facilitating Podcast Episode Browsing with Auto-generated Chapters](https://mlsanity.com/paper/2410.16148/) (2024)
- [An Interactive Query Generation Assistant using LLM-based Prompt Modification and User Feedback](https://mlsanity.com/paper/2311.11226/) (2023)
- [Database-Augmented Query Representation for Information Retrieval](https://mlsanity.com/paper/2406.16013/) (2024)
- [RL-Index: Reinforcement Learning for Retrieval Index Reasoning](https://mlsanity.com/paper/2606.16316/) (2026)
- [LLM-based IR-system for Bank Supervisors](https://mlsanity.com/paper/2508.02945/) (2025)
- [A Benchmark for Text Expansion: Datasets, Metrics, and Baselines](https://mlsanity.com/paper/2309.09198/) (2023)
- [Chunk Knowledge Generation Model for Enhanced Information Retrieval: A Multi-task Learning Approach](https://mlsanity.com/paper/2509.15658/) (2025)

## Abstract (arXiv)

One of the challenges in information retrieval (IR) is the vocabulary mismatch problem, which happens when the terms between queries and documents are lexically different but semantically similar. While recent work has proposed to expand the queries or documents by enriching their representations with additional relevant terms to address this challenge, they usually require a large volume of query-document pairs to train an expansion model. In this paper, we propose an Unsupervised Document Expansion with Generation (UDEG) framework with a pre-trained language model, which generates diverse supplementary sentences for the original document without using labels on query-document pairs for training. For generating sentences, we further stochastically perturb their embeddings to generate more diverse sentences for document expansion. We validate our framework on two standard IR benchmark datasets. The results show that our framework significantly outperforms relevant expansion baselines for IR.