A White Box Analysis of ColBERT - Sorbonne Université Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2020

A White Box Analysis of ColBERT

Résumé

Transformer-based models are nowadays state-of-the-art in ad-hoc Information Retrieval, but their behavior is far from being understood. Recent work has claimed that BERT does not satisfy the classical IR axioms. However, we propose to dissect the matching process of ColBERT, through the analysis of term importance and exact/soft matching patterns. Even if the traditional axioms are not formally verified, our analysis reveals that ColBERT: (i) is able to capture a notion of term importance; (ii) relies on exact matches for important terms.

Dates et versions

hal-03084279 , version 1 (21-12-2020)

Identifiants

Citer

Thibault Formal, Benjamin Piwowarski, Stéphane Clinchant. A White Box Analysis of ColBERT. 2020. ⟨hal-03084279⟩
21 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More