Mind the Labels: Describing Relations in Knowledge Graphs With Pretrained Models

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F23%3A10475957" target="_blank" >RIV/00216208:11320/23:10475957 - isvavai.cz</a>

Výsledek na webu

<a href="https://aclanthology.org/2023.eacl-main.176/" target="_blank" >https://aclanthology.org/2023.eacl-main.176/</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.18653/v1/2023.eacl-main.176" target="_blank" >10.18653/v1/2023.eacl-main.176</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Mind the Labels: Describing Relations in Knowledge Graphs With Pretrained Models

Popis výsledku v původním jazyce

Pretrained language models (PLMs) for data-to-text (D2T) generation can use human-readable data labels such as column headings, keys, or relation names to generalize to out-of-domain examples. However, the models are well-known in producing semantically inaccurate outputs if these labels are ambiguous or incomplete, which is often the case in D2T datasets. In this paper, we expose this issue on the task of descibing a relation between two entities. For our experiments, we collect a novel dataset for verbalizing a diverse set of 1,522 unique relations from three large-scale knowledge graphs (Wikidata, DBPedia, YAGO). We find that although PLMs for D2T generation expectedly fail on unclear cases, models trained with a large variety of relation labels are surprisingly robust in verbalizing novel, unseen relations. We argue that using data with a diverse set of clear and meaningful labels is key to training D2T generation systems capable of generalizing to novel domains.

Název v anglickém jazyce

Mind the Labels: Describing Relations in Knowledge Graphs With Pretrained Models

Popis výsledku anglicky

Pretrained language models (PLMs) for data-to-text (D2T) generation can use human-readable data labels such as column headings, keys, or relation names to generalize to out-of-domain examples. However, the models are well-known in producing semantically inaccurate outputs if these labels are ambiguous or incomplete, which is often the case in D2T datasets. In this paper, we expose this issue on the task of descibing a relation between two entities. For our experiments, we collect a novel dataset for verbalizing a diverse set of 1,522 unique relations from three large-scale knowledge graphs (Wikidata, DBPedia, YAGO). We find that although PLMs for D2T generation expectedly fail on unclear cases, models trained with a large variety of relation labels are surprisingly robust in verbalizing novel, unseen relations. We argue that using data with a diverse set of clear and meaningful labels is key to training D2T generation systems capable of generalizing to novel domains.

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

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2023

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics

ISBN

978-1-959429-44-9

ISSN

—

e-ISSN

—

Počet stran výsledku

18

Strana od-do

2398-2415

Název nakladatele

Association for Computational Linguistics

Místo vydání

Stroudsburg, PA, USA

Místo konání akce

Dubrovnik, Croatia

Datum konání akce

2. 5. 2023

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—