Learning To Retrieve: How to Train a Dense Retrieval Model Effectively and Efficiently

Abstract

Ranking has always been one of the top concerns in information retrieval research. For decades, lexical matching signal has dominated the ad-hoc retrieval process, but it also has inherent defects, such as the vocabulary mismatch problem. Recently, Dense Retrieval (DR) technique has been proposed to alleviate these limitations by capturing the deep semantic relationship between queries and documents. The training of most existing Dense Retrieval models relies on sampling negative instances from the corpus to optimize a pairwise loss function. Through investigation, we find that this kind of training strategy is biased and fails to optimize full retrieval performance effectively and efficiently. To solve this problem, we propose a Learning To Retrieve (LTRe) training technique. LTRe constructs the document index beforehand. At each training iteration, it performs full retrieval without negative sampling and then updates the query representation model parameters. Through this process, it teaches the DR model how to retrieve relevant documents from the entire corpus instead of how to rerank a potentially biased sample of documents. Experiments in both passage retrieval and document retrieval tasks show that: 1) in terms of effectiveness, LTRe significantly outperforms all competitive sparse and dense baselines. It even gains better performance than the BM25-BERT cascade system under reasonable latency constraints. 2) in terms of training efficiency, compared with the previous state-of-the-art DR method, LTRe provides more than 170x speed-up in the training process. Training with a compressed index further saves computing resources with minor performance loss.

References in corpus (17)

• Decoupled Weight Decay Regularization
• Longformer: The Long-Document Transformer
• Convolutional Neural Network Architectures for Matching Natural Language Sentences
• REALM: Retrieval-Augmented Language Model Pre-Training
• Deeper Text Understanding for IR with Contextual Neural Language Modeling
• Passage Re-ranking with BERT
• Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval
• Document Expansion by Query Prediction
• Multi-Stage Document Ranking with BERT
• A Deep Architecture for Semantic Matching with Multiple Positional Sentence Representations
• Large Batch Optimization for Deep Learning: Training BERT in 76 minutes
• Context-Aware Sentence/Passage Term Importance Estimation For First Stage Retrieval
• Sparse, Dense, and Attentional Representations for Text Retrieval
• Learning-to-Rank with BERT in TF-Ranking
• Complementing Lexical Retrieval with Semantic Residual Embedding
• RepBERT: Contextualized Text Embeddings for First-Stage Retrieval
• Overview of the TREC 2021 deep learning track

Cited by in corpus (5)

• Information Retrieval: Recent Advances and Beyond
• CODER: An efficient framework for improving retrieval through COntextual Document Embedding Reranking
• Building an Efficient and Effective Retrieval-based Dialogue System via Mutual Learning
• Improving Query Representations for Dense Retrieval with Pseudo Relevance Feedback
• Jointly Optimizing Query Encoder and Product Quantization to Improve Retrieval Performance