Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval

By Lee Xiong, Chenyan Xiong, Ye Li, Kwok-Fung Tang, Jialin Liu, Paul N. Bennett, Junaid Ahmed, Arnold Overwijk. 2020

1,631 citations recorded by the source corpus.

Conducting text retrieval in a dense learned representation space has many intriguing advantages over sparse retrieval. Yet the effectiveness of dense retrieval (DR) often requires combination with sparse retrieval. In this paper, we identify that the main bottleneck is in the training mechanisms, where the negative instances used in training are not representative of the irrelevant documents in testing. This paper presents Approximate nearest neighbor Negative Contrastive Estimation (ANCE), a training mechanism that constructs negatives from an Approximate Nearest Neighbor (ANN) index of the corpus, which is parallelly updated with the learning process to select more realistic negative training instances. This fundamentally resolves the discrepancy between the data distribution used in the training and testing of DR. In our experiments, ANCE boosts the BERT-Siamese DR model to outperform all competitive dense and sparse retrieval baselines. It nearly matches the accuracy of sparse-retrieval-and-BERT-reranking using dot-product in the ANCE-learned representation space and provides almost 100x speed-up.

Subfields: Artificial Intelligence, Information Retrieval & Search

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

  1. Luke S. Zettlemoyer
  2. Mohit Bansal
  3. Jianfeng Gao
  4. Hannaneh Hajishirzi
  5. Jimmy J. Lin
  6. Graham Neubig
  7. Percy Liang
  8. Sebastian Riedel

In Information Retrieval & Search

  1. Jimmy J. Lin
  2. Jamie Callan
  3. Luyu Gao
  4. Sebastian Hofstätter
  5. Wen-tau Yih
  6. Sebastian Riedel
  7. Omar Khattab
  8. Gautier Izacard

Citations in the corpus