Abstract
Publications home
Match your words! A study of lexical matching in neural information retrieval
at
| Thibault Formal, Benjamin Piwowarski, Stéphane Clinchant |
| 44th European Conference on Information Retrieval (ECIR), Stavanger, Norway, 10-14 April, 2022 |
| arXiv |
Neural Information Retrieval models hold the promise to replace lexical matching models, e.g. BM25, in modern search engines. While their capabilities have fully shone on in-domain datasets like MS MARCO, they have recently been challenged on out-of-domain zero-shot settings (BEIR benchmark), questioning their actual generalization capabilities compared to bag-of-words approaches. Particularly, we wonder if these shortcomings are (partly) due to the inability of neural IR models to perform lexical matching off-the-shelf. We first define how to measure the discrepancy between the lexical matching performed by the model and an “ideal” one. Based on this, we study the behavior of different state-of-the-art neural IR models, more particularly focusing on whether they are able to perform lexical matching \emph{when it’s actually useful}. Overall, we show that neural IR models fail to properly generalize term importance on out-of-domain collections or terms unseen during training.
Related Content
This web site uses cookies for the site search, to display videos and for aggregate site analytics.
Learn more about these cookies in our privacy notice.
Cookie settings
You may choose which kind of cookies you allow when visiting this website. Click on "Save cookie settings" to apply your choice.
This content is currently blocked. To view the content please either 'Accept social media cookies' or 'Accept all cookies'.
For more information on cookies see our privacy notice.