CorPipe 24

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F24%3A10493028" target="_blank" >RIV/00216208:11320/24:10493028 - isvavai.cz</a>

Výsledek na webu

<a href="https://github.com/ufal/crac2024-corpipe" target="_blank" >https://github.com/ufal/crac2024-corpipe</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

CorPipe 24

Popis výsledku v původním jazyce

We present CorPipe 24, the winning entry to the CRAC 2024 Shared Task on Multilingual Coreference Resolution. In this third iteration of the shared task, a novel objective is to also predict empty nodes needed for zero coreference mentions (while the empty nodes were given on input in previous years). This way, coreference resolution can be performed on raw text. We evaluate two model variants: a two-stage approach (where the empty nodes are predicted first using a pretrained encoder model and then processed together with sentence words by another pretrained model) and a single-stage approach (where a single pretrained encoder model generates empty nodes, coreference mentions, and coreference links jointly). In both settings, CorPipe surpasses other participants by a large margin of 3.9 and 2.8 percent points, respectively. The source code and the trained model are available at https://github.com/ufal/crac2024-corpipe.

Název v anglickém jazyce

CorPipe 24

Popis výsledku anglicky

We present CorPipe 24, the winning entry to the CRAC 2024 Shared Task on Multilingual Coreference Resolution. In this third iteration of the shared task, a novel objective is to also predict empty nodes needed for zero coreference mentions (while the empty nodes were given on input in previous years). This way, coreference resolution can be performed on raw text. We evaluate two model variants: a two-stage approach (where the empty nodes are predicted first using a pretrained encoder model and then processed together with sentence words by another pretrained model) and a single-stage approach (where a single pretrained encoder model generates empty nodes, coreference mentions, and coreference links jointly). In both settings, CorPipe surpasses other participants by a large margin of 3.9 and 2.8 percent points, respectively. The source code and the trained model are available at https://github.com/ufal/crac2024-corpipe.

Druh

R - Software

CEP obor

—

OECD FORD obor

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

Projekt

<a href="/cs/project/GX20-16819X" target="_blank" >GX20-16819X: Porozumění jazyku: od syntaxe k diskurzu</a><br>

Návaznosti

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

Rok uplatnění

2024

Kód důvěrnosti údajů

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

Interní identifikační kód produktu

[https://github.com/ufal/crac202

Technické parametry

Výsledek volně dostupný na adrese https://github.com/ufal/crac2024-corpipe.

Ekonomické parametry

146419

IČO vlastníka výsledku

00216208

Název vlastníka

Univerzita Karlova