Learning compositional structures for semantic graph parsing

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F21%3A10440949" target="_blank" >RIV/00216208:11320/21:10440949 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Learning compositional structures for semantic graph parsing

Popis výsledku v původním jazyce

AM dependency parsing is a method for neural semantic graph parsing that exploits the principle of compositionality. While AM dependency parsers have been shown to be fast and accurate across several graphbanks, they require explicit annotations of the compositional tree structures for training. In the past, these were obtained using complex graphbank-specific heuristics written by experts. Here we show how they can instead be trained directly on the graphs with a neural latent-variable model, drastically reducing the amount and complexity of manual heuristics. We demonstrate that our model picks up on several linguistic phenomena on its own and achieves comparable accuracy to supervised training, greatly facilitating the use of AM dependency parsing for new sembanks.

Název v anglickém jazyce

Learning compositional structures for semantic graph parsing

Popis výsledku anglicky

AM dependency parsing is a method for neural semantic graph parsing that exploits the principle of compositionality. While AM dependency parsers have been shown to be fast and accurate across several graphbanks, they require explicit annotations of the compositional tree structures for training. In the past, these were obtained using complex graphbank-specific heuristics written by experts. Here we show how they can instead be trained directly on the graphs with a neural latent-variable model, drastically reducing the amount and complexity of manual heuristics. We demonstrate that our model picks up on several linguistic phenomena on its own and achieves comparable accuracy to supervised training, greatly facilitating the use of AM dependency parsing for new sembanks.

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

2021

Kód důvěrnosti údajů

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

Název statě ve sborníku

SPNLP 2021: THE 5TH WORKSHOP ON STRUCTURED PREDICTION FOR NLP

ISBN

978-1-954085-75-6

ISSN

—

e-ISSN

—

Počet stran výsledku

11

Strana od-do

22-32

Název nakladatele

ASSOC COMPUTATIONAL LINGUISTICS-ACL

Místo vydání

STROUDSBURG

Místo konání akce

online

Datum konání akce

6. 8. 2021

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000694721100003