A White Box Analysis of ColBERT - Archive ouverte HAL Access content directly
Conference Papers Year : 2021

A White Box Analysis of ColBERT

Abstract

Transformer-based models are nowadays state-of-the-art in adhoc Information Retrieval, but their behavior are 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.
Fichier principal
Vignette du fichier
Formal et al_2020_A White Box Analysis of ColBERT.pdf (474.45 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03364396 , version 1 (07-10-2021)

Identifiers

Cite

Thibault Formal, Benjamin Piwowarski, Stéphane Clinchant. A White Box Analysis of ColBERT. 43rd EUROPEAN CONFERENCE ON INFORMATION RETRIEVAL, Mar 2021, Lucca (online), Italy. pp.257-263, ⟨10.1007/978-3-030-72240-1_23⟩. ⟨hal-03364396⟩
35 View
33 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More