The Functional Relevance of Probed Information: A Case Study

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F23%3A10475962" target="_blank" >RIV/00216208:11320/23:10475962 - isvavai.cz</a>

Result on the web

<a href="https://aclanthology.org/2023.eacl-main.58" target="_blank" >https://aclanthology.org/2023.eacl-main.58</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

The Functional Relevance of Probed Information: A Case Study

Original language description

Recent studies have shown that transformer models like BERT rely on number information encoded in their representations of sentences&apos; subjects and head verbs when performing subject-verb agreement. However, probing experiments suggest that subject number is also encoded in the representations of all words in such sentences. In this paper, we use causal interventions to show that BERT only uses the subject plurality information encoded in its representations of the subject and words that agree with it in number. We also demonstrate that current probing metrics are unable to determine which words&apos; representations contain functionally relevant information. This both provides a revised view of subject-verb agreement in language models, and suggests potential pitfalls for current probe usage and evaluation.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

<a href="/en/project/GA23-06912S" target="_blank" >GA23-06912S: Identification and Prevention of Unwanted Gender Bias in Neural Language Models</a><br>

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2023

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics

ISBN

978-1-959429-44-9

ISSN

—

e-ISSN

—

Number of pages

14

Pages from-to

835-848

Publisher name

Association for Computational Linguistics

Place of publication

Stroudsburg, PA, USA

Event location

Dubrovnik, Croatia

Event date

May 2, 2023

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—