Transformer Memory as a Differentiable Search Index

Abstract

In this paper, we demonstrate that information retrieval can be accomplished with a single Transformer, in which all information about the corpus is encoded in the parameters of the model. To this end, we introduce the Differentiable Search Index (DSI), a new paradigm that learns a text-to-text model that maps string queries directly to relevant docids; in other words, a DSI model answers queries directly using only its parameters, dramatically simplifying the whole retrieval process. We study variations in how documents and their identifiers are represented, variations in training procedures, and the interplay between models and corpus sizes. Experiments demonstrate that given appropriate design choices, DSI significantly outperforms strong baselines such as dual encoder models. Moreover, DSI demonstrates strong generalization capabilities, outperforming a BM25 baseline in a zero-shot setup.

NeurIPS 2022

Cited by in corpus (8)

• Large Language Models as Zero-Shot Conversational Recommenders
• Advances of Machine Learning in Materials Science: Ideas and Techniques
• CorpusBrain: Pre-train a Generative Retrieval Model for Knowledge-Intensive Language Tasks
• Continual Learning for Generative Retrieval over Dynamic Corpora
• Content-Based Collaborative Generation for Recommender Systems
• MS MARCO Web Search: a Large-scale Information-rich Web Dataset with Millions of Real Click Labels
• PEFA: Parameter-Free Adapters for Large-scale Embedding-based Retrieval Models
• De-DSI: Decentralised Differentiable Search Index