MELA: Multilingual Evaluation of Linguistic Acceptability

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F25%3AVDHW2J8W" target="_blank" >RIV/00216208:11320/25:VDHW2J8W - isvavai.cz</a>

Výsledek na webu

<a href="https://www.scopus.com/inward/record.uri?eid=2-s2.0-85204436622&partnerID=40&md5=936049d43bab424e20b243a2baa4b851" target="_blank" >https://www.scopus.com/inward/record.uri?eid=2-s2.0-85204436622&partnerID=40&md5=936049d43bab424e20b243a2baa4b851</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

MELA: Multilingual Evaluation of Linguistic Acceptability

Popis výsledku v původním jazyce

In this work, we present the largest benchmark to date on linguistic acceptability: Multilingual Evaluation of Linguistic Acceptability-MELA, with 46K samples covering 10 languages from a diverse set of language families. We establish LLM baselines on this benchmark, and investigate cross-lingual transfer in acceptability judgements with XLM-R. In pursuit of multilingual interpretability, we conduct probing experiments with fine-tuned XLM-R to explore the process of syntax capability acquisition. Our results show that GPT-4o exhibits a strong multilingual ability, outperforming fine-tuned XLM-R, while open-source multilingual models lag behind by a noticeable gap. Cross-lingual transfer experiments show that transfer in acceptability judgment is non-trivial: 500 Icelandic fine-tuning examples lead to 23 MCC performance in a completely unrelated language-Chinese. Results of our probing experiments indicate that training on MELA improves the performance of XLM-R on syntax-related tasks. https://github.com/sjtu-compling/MELA. © 2024 Association for Computational Linguistics.

Název v anglickém jazyce

MELA: Multilingual Evaluation of Linguistic Acceptability

Popis výsledku anglicky

In this work, we present the largest benchmark to date on linguistic acceptability: Multilingual Evaluation of Linguistic Acceptability-MELA, with 46K samples covering 10 languages from a diverse set of language families. We establish LLM baselines on this benchmark, and investigate cross-lingual transfer in acceptability judgements with XLM-R. In pursuit of multilingual interpretability, we conduct probing experiments with fine-tuned XLM-R to explore the process of syntax capability acquisition. Our results show that GPT-4o exhibits a strong multilingual ability, outperforming fine-tuned XLM-R, while open-source multilingual models lag behind by a noticeable gap. Cross-lingual transfer experiments show that transfer in acceptability judgment is non-trivial: 500 Icelandic fine-tuning examples lead to 23 MCC performance in a completely unrelated language-Chinese. Results of our probing experiments indicate that training on MELA improves the performance of XLM-R on syntax-related tasks. https://github.com/sjtu-compling/MELA. © 2024 Association for Computational Linguistics.

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

2024

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proc. Annu. Meet. Assoc. Comput Linguist.

ISBN

979-889176094-3

ISSN

0736-587X

e-ISSN

—

Počet stran výsledku

17

Strana od-do

2658-2674

Název nakladatele

Association for Computational Linguistics (ACL)

Místo vydání

—

Místo konání akce

Bangkok

Datum konání akce

1. 1. 2025

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—