NAISS
SUPR
NAISS Projects
SUPR
Toward Visually Grounded MCTS for Visual Reasoning
Dnr:

NAISS 2026/4-1222

Type:

NAISS Small

Principal Investigator:

Aron Henriksson

Affiliation:

Stockholms universitet

Start Date:

2026-06-29

End Date:

2027-07-01

Primary Classification:

10208: Natural Language Processing

Webpage:

Allocation

Abstract

Vision-language models (VLMs) have achieved strong performance in multimodal benchmarks, but they often rely on learned language priors rather than the actual image. The ViLP (Probing Visual Language Priors) benchmark specifically exposes this failure case by including strong distractor facts in the questions, in order to trick the model to answer based on its linguistic knowledge, without consulting the image. So, despite near-perfect human performance, modern VLMs show low performance. This project aims to develop novel methods for visually grounded reasoning in VLMs.