Learning Diverse Document Representations with Deep Query Interactions for Dense Retrieval

Zehan Li, Nan Yang, Liang Wang, Furu Wei
8/8/2022
cs.IRcs.CL

Official code: jordane95/dual-cross-encoder (17 stars, Apache-2.0, last commit 2023-03-15).

Abstract

In this paper, we propose a new dense retrieval model which learns diverse document representations with deep query interactions. Our model encodes each document with a set of generated pseudo-queries to get query-informed, multi-view document representations. It not only enjoys high inference efficiency like the vanilla dual-encoder models, but also enables deep query-document interactions in document encoding and provides multi-faceted representations to better match different queries. Experiments on several benchmarks demonstrate the effectiveness of the proposed method, out-performing strong dual encoder baselines.The code is available at \url{https://github.com/jordane95/dual-cross-encoder

View on arXivView PDF

Code Implementations(1)

jordane95/dual-cross-encoderOfficial100%
Apache-2.0

Dual Cross Encoder for Dense Retrieval

172Shell, PythonAug 2, 20223 years agoApache-2.0
cross-encoderdense-retrievaldual-encoderquery-generation

Similar Papers

SPENCER: Self-Adaptive Model Distillation for Efficient Code Retrieval

84% match2025

Quick Dense Retrievers Consume KALE: Post Training Kullback Leibler Alignment of Embeddings for Asymmetrical dual encoders

84% match2023

A Symmetric Dual Encoding Dense Retrieval Framework for Knowledge-Intensive Visual Question Answering

82% match1 repo142023

Unsupervised Text Representation Learning via Instruction-Tuning for Zero-Shot Dense Retrieval

82% match2024

ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT

82% match1 repo3.9k2020

How to Train Your DRAGON: Diverse Augmentation Towards Generalizable Dense Retrieval

82% match1 repo2962023

Cite this paper

@article{li2022learning,
  title  = {Learning Diverse Document Representations with Deep Query Interactions for Dense Retrieval},
  author = {Zehan Li and Nan Yang and Liang Wang and Furu Wei},
  year   = {2022},
  eprint = {2208.04232},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2208.04232v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2208.04232.svg)](https://codeofpaper.com/paper/2208.04232)

Discussion