Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Tan, Alvin Wei Ming, Prystawski, Ben, Boyce, Veronica, Frank, Michael C.
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2511.03908
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909889480097792
author Tan, Alvin Wei Ming
Prystawski, Ben
Boyce, Veronica
Frank, Michael C.
author_facet Tan, Alvin Wei Ming
Prystawski, Ben
Boyce, Veronica
Frank, Michael C.
contents Iterated reference games - in which players repeatedly pick out novel referents using language - present a test case for agents' ability to perform context-sensitive pragmatic reasoning in multi-turn linguistic environments. We tested humans and vision-language models on trials from iterated reference games, varying the given context in terms of amount, order, and relevance. Without relevant context, models were above chance but substantially worse than humans. However, with relevant context, model performance increased dramatically over trials. Few-shot reference games with abstract referents remain a difficult task for machine learning models.
format Preprint
id arxiv_https___arxiv_org_abs_2511_03908
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Context informs pragmatic interpretation in vision-language models
Tan, Alvin Wei Ming
Prystawski, Ben
Boyce, Veronica
Frank, Michael C.
Computation and Language
Iterated reference games - in which players repeatedly pick out novel referents using language - present a test case for agents' ability to perform context-sensitive pragmatic reasoning in multi-turn linguistic environments. We tested humans and vision-language models on trials from iterated reference games, varying the given context in terms of amount, order, and relevance. Without relevant context, models were above chance but substantially worse than humans. However, with relevant context, model performance increased dramatically over trials. Few-shot reference games with abstract referents remain a difficult task for machine learning models.
title Context informs pragmatic interpretation in vision-language models
topic Computation and Language
url https://arxiv.org/abs/2511.03908