Efficient Passage Retrieval with Hashing for Open-domain Question Answering

By Ikuya Yamada, Akari Asai, Hannaneh Hajishirzi. 2021

99 citations recorded by the source corpus.

Most state-of-the-art open-domain question answering systems use a neural retrieval model to encode passages into continuous vectors and extract them from a knowledge source. However, such retrieval models often require large memory to run because of the massive size of their passage index. In this paper, we introduce Binary Passage Retriever (BPR), a memory-efficient neural retrieval model that integrates a learning-to-hash technique into the state-of-the-art Dense Passage Retriever (DPR) to represent the passage index using compact binary codes rather than continuous vectors. BPR is trained with a multi-task objective over two tasks: efficient candidate generation based on binary codes and accurate reranking based on continuous vectors. Compared with DPR, BPR substantially reduces the memory cost from 65GB to 2GB without a loss of accuracy on two standard open-domain question answering benchmarks: Natural Questions and TriviaQA. Our code and trained models are available at https://github.com/studio-ousia/bpr.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, 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

In Computer Vision and Pattern Recognition

In Information Retrieval & Search

Citations in the corpus

Autoregressive Search Engines: Generating Substrings as Document Identifiers (2022, background): “We report in index size, this still comes at the expense of performance (Yamada et al., 2021; Lewis et al., 2021a).”
The Web Is Your Oyster - Knowledge-Intensive NLP against a Very Large Web Corpus (2021, methodology): “phisticated bi-encoder training pipelines (Yamada et al., 2021; Zhan et al., 2021).”
Dense Text Retrieval Based on Pretrained Language Models: A Survey (2022, background): “[264] incorporate a learned hash function to represent both queries and texts as binary codes in an end-to-end way.”
ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interaction (2021, background): “The Binary Passage Retriever (BPR; Yamada et al. 2021) learns to directly hash embeddings to binary codes using a differentiable tanh function.”
Task-aware Retrieval with Instructions (2022, methodology): “For this model, document embeddings can be computed offline, improving inference efficiency at the cost of storage space (Yamada et al., 2021).”
A Few Brief Notes on DeepImpact, COIL, and a Conceptual Framework for Information Retrieval Techniques (2021, unknown): “We note, however, that with dense retrieval techniques, ﬁxed-width vectors can be approximated with binary hash codes, yielding far more compact representations with sacriﬁcing much effectiveness (Yamada et al., 2021).”
Boosted Dense Retriever (2021, background): “We also show that DrBoost’s index is amenable to compression, and can be compressed to 800MB, 2.5x smaller than a recent state of the art efﬁcient retriever, whilst being more accurate (Yamada et al., 2021).”
HAKARI-Bench: A Lightweight Benchmark for Comparing Retrieval Architectures and Efficiency Settings under Unified Conditions (2026, unknown): “In particular, the two-stage configuration of “generating candidates efficiently with binary codes and re-ranking (rescoring) accurately with continuous vectors” traces back to the Binary Passage Retriever (Yamada et al., 2021).”