English

SimLM: Pre-training with Representation Bottleneck for Dense Passage Retrieval

Information Retrieval 2023-05-15 v2

Abstract

In this paper, we propose SimLM (Similarity matching with Language Model pre-training), a simple yet effective pre-training method for dense passage retrieval. It employs a simple bottleneck architecture that learns to compress the passage information into a dense vector through self-supervised pre-training. We use a replaced language modeling objective, which is inspired by ELECTRA, to improve the sample efficiency and reduce the mismatch of the input distribution between pre-training and fine-tuning. SimLM only requires access to unlabeled corpus, and is more broadly applicable when there are no labeled data or queries. We conduct experiments on several large-scale passage retrieval datasets, and show substantial improvements over strong baselines under various settings. Remarkably, SimLM even outperforms multi-vector approaches such as ColBERTv2 which incurs significantly more storage cost. Our code and model check points are available at https://github.com/microsoft/unilm/tree/master/simlm .

Keywords

Cite

@article{arxiv.2207.02578,
  title  = {SimLM: Pre-training with Representation Bottleneck for Dense Passage Retrieval},
  author = {Liang Wang and Nan Yang and Xiaolong Huang and Binxing Jiao and Linjun Yang and Daxin Jiang and Rangan Majumder and Furu Wei},
  journal= {arXiv preprint arXiv:2207.02578},
  year   = {2023}
}

Comments

Accepted to ACL 2023

R2 v1 2026-06-24T12:15:42.069Z