VLMs using language-guided inference capture context-sensitivity of human object recognition behavior

Human vision is a context-sensitive process that interprets objects in relation to their surroundings. While behavioral research has long shown that scene context facilitates object recognition, the underlying computational mechanisms—and the extent to which artificial vision models replicate this a...

Ausführliche Beschreibung

Bibliografische Detailangaben

Hauptverfasser:	Karim Rajaei, Radoslaw Martin Cichy, Hamid Soltanian-Zadeh
Format:	Artigo
Sprache:	Inglês
Veröffentlicht:	Elsevier 2026-03-01
Schriftenreihe:	Array
Schlagworte:	Contextual object recognition Vision-language models (VLMs) Language supervision Language-guided inference Vision transformers (ViTs) Human–AI comparison
Online-Zugang:	http://www.sciencedirect.com/science/article/pii/S2590005626000548
Tags:	Tag hinzufügen Keine Tags, Fügen Sie das erste Tag hinzu!