Sparse, Dense, and Attentional Representations for Text Retrieval
AbstractDual encoders perform retrieval by encoding documents and queries into dense low-dimensional vectors, scoring each document by its inner product with the query. We investigate the capacity of this architecture relative to sparse bag-of-words models and attentional neural netw...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
The MIT Press
2021-01-01
|
Series: | Transactions of the Association for Computational Linguistics |
Online Access: | https://direct.mit.edu/tacl/article/doi/10.1162/tacl_a_00369/100684/Sparse-Dense-and-Attentional-Representations-for |