Where Does Linguistic Information Emerge in Neural Language Models? Measuring Gains and Contributions across Layers

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F22%3A8T27XUX5" target="_blank" >RIV/00216208:11320/22:8T27XUX5 - isvavai.cz</a>

Výsledek na webu

<a href="https://aclanthology.org/2022.coling-1.413" target="_blank" >https://aclanthology.org/2022.coling-1.413</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Where Does Linguistic Information Emerge in Neural Language Models? Measuring Gains and Contributions across Layers

Popis výsledku v původním jazyce

Probing studies have extensively explored where in neural language models linguistic information is located. The standard approach to interpreting the results of a probing classifier is to focus on the layers whose representations give the highest performance on the probing task. We propose an alternative method that asks where the task-relevant information emerges in the model. Our framework consists of a family of metrics that explicitly model local information gain relative to the previous layer and each layer's contribution to the model's overall performance. We apply the new metrics to two pairs of syntactic probing tasks with different degrees of complexity and find that the metrics confirm the expected ordering only for one of the pairs. Our local metrics show a massive dominance of the first layers, indicating that the features that contribute the most to our probing tasks are not as high-level as global metrics suggest.

Název v anglickém jazyce

Where Does Linguistic Information Emerge in Neural Language Models? Measuring Gains and Contributions across Layers

Popis výsledku anglicky

Probing studies have extensively explored where in neural language models linguistic information is located. The standard approach to interpreting the results of a probing classifier is to focus on the layers whose representations give the highest performance on the probing task. We propose an alternative method that asks where the task-relevant information emerges in the model. Our framework consists of a family of metrics that explicitly model local information gain relative to the previous layer and each layer's contribution to the model's overall performance. We apply the new metrics to two pairs of syntactic probing tasks with different degrees of complexity and find that the metrics confirm the expected ordering only for one of the pairs. Our local metrics show a massive dominance of the first layers, indicating that the features that contribute the most to our probing tasks are not as high-level as global metrics suggest.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

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

Projekt

—

Návaznosti

—

Rok uplatnění

2022

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proceedings of the 29th International Conference on Computational Linguistics

ISBN

—

ISSN

2951-2093

e-ISSN

—

Počet stran výsledku

13

Strana od-do

4664-4676

Název nakladatele

International Committee on Computational Linguistics

Místo vydání

—

Místo konání akce

Gyeongju, Republic of Korea

Datum konání akce

1. 1. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—