0

Unsupervised Corpus Aware Language Model Pre-training for Dense Passage Retrieval

coCondenser, an improvement to the Condenser pre-training architecture, enhances dense retrieval by adding a contrastive loss, reducing the need for heavy data engineering and large batch sizes.

Year
2021
Venue
ACL 2022 5
Authors
2
Hosting
Abstract onlyARXIV-DEFAULT

Cite

Notes

Only stored in your browser.

Attribution

Abstract & full text
arxiv.org/abs/2108.05540ARXIV-DEFAULT
TL;DR
Semantic Scholar
Attribution policy →

Abstract

Recent research demonstrates the effectiveness of using fine-tuned language models~(LM) for dense retrieval. However, dense retrievers are hard to train, typically requiring heavily engineered fine-tuning pipelines to realize their full potential. In this paper, we identify and address two underlying problems of dense retrievers: i)~fragility to training data noise and ii)~requiring large batches to robustly learn the embedding space. We use the recently proposed Condenser pre-training architecture, which learns to condense information into the dense vector through LM pre-training. On top of it, we propose coCondenser, which adds an unsupervised corpus-level contrastive loss to warm up the passage embedding space. Retrieval experiments on MS-MARCO, Natural Question, and Trivia QA datasets show that coCondenser removes the need for heavy data engineering such as augmentation, synthesis, or filtering, as well as the need for large batch training. It shows comparable performance to RocketQA, a state-of-the-art, heavily engineered system, using simple small batch fine-tuning.

Authors

2