A Game Interface to Study Semantic Grounding in Text-Based Models - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

A Game Interface to Study Semantic Grounding in Text-Based Models

Résumé

Can language models learn grounded representations from text distribution alone? This question is both central and recurrent in natural language processing; authors generally agree that grounding requires more than textual distribution. We propose to experimentally test this claim: if any two words have different meanings and yet cannot be distinguished from distribution alone, then grounding is out of the reach of textbased models. To that end, we present early work on an online game for the collection of human judgments on the distributional similarity of word pairs in five languages. We further report early results of our data collection campaign.
Fichier principal
Vignette du fichier
game_interface_grounding_mickus_et_al.pdf (375.62 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03547413 , version 1 (28-01-2022)

Identifiants

  • HAL Id : hal-03547413 , version 1

Citer

Timothee Mickus, Mathieu Constant, Denis Paperno. A Game Interface to Study Semantic Grounding in Text-Based Models. 2021 IEEE Conference on Games, Aug 2021, Copenhagen, Denmark. ⟨hal-03547413⟩
37 Consultations
27 Téléchargements

Partager

Gmail Facebook X LinkedIn More