Designing a strong test for measuring true common-sense reasoning

Designing a strong test for measuring true common-sense reasoning

Common-sense reasoning has recently emerged as an important test for artificial general intelligence, especially given the much-publicized successes of language representation models such as T5, BERT and GPT-3. Currently, typical benchmarks involve question answering tasks, but to test the full comp...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Nature machine intelligence 2022-04, Vol.4 (4), p.318-322
Hauptverfasser: Kejriwal, Mayank, Santos, Henrique, Mulvehill, Alice M., McGuinness, Deborah L.
Format: Artikel
Sprache:eng
Schlagworte:
639/705/1042
639/705/117
Artificial intelligence
Cognition & reasoning
Comment
Engineering
Language
Natural language
Physics
Reasoning
Science
Task complexity
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Common-sense reasoning has recently emerged as an important test for artificial general intelligence, especially given the much-publicized successes of language representation models such as T5, BERT and GPT-3. Currently, typical benchmarks involve question answering tasks, but to test the full complexity of common-sense reasoning, more comprehensive evaluation methods that are grounded in theory should be developed.
ISSN:2522-5839
2522-5839
DOI:10.1038/s42256-022-00478-4