Transferring Neural Potentials For High Order Dependency Parsing

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F23%3AGK8FGIEG" target="_blank" >RIV/00216208:11320/23:GK8FGIEG - isvavai.cz</a>

Výsledek na webu

<a href="http://arxiv.org/abs/2306.10469" target="_blank" >http://arxiv.org/abs/2306.10469</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Transferring Neural Potentials For High Order Dependency Parsing

Popis výsledku v původním jazyce

"High order dependency parsing leverages high order features such as siblings or grandchildren to improve state of the art accuracy of current first order dependency parsers. The present paper uses biaffine scores to provide an estimate of the arc scores and is then propagated into a graphical model. The inference inside the graphical model is solved using dual decomposition. The present algorithm propagates biaffine neural scores to the graphical model and by leveraging dual decomposition inference, the overall circuit is trained end-to-end to transfer first order informations to the high order informations."

Název v anglickém jazyce

Transferring Neural Potentials For High Order Dependency Parsing

Popis výsledku anglicky

"High order dependency parsing leverages high order features such as siblings or grandchildren to improve state of the art accuracy of current first order dependency parsers. The present paper uses biaffine scores to provide an estimate of the arc scores and is then propagated into a graphical model. The inference inside the graphical model is solved using dual decomposition. The present algorithm propagates biaffine neural scores to the graphical model and by leveraging dual decomposition inference, the overall circuit is trained end-to-end to transfer first order informations to the high order informations."

Druh

O - Ostatní výsledky

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

2023

Kód důvěrnosti údajů

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