A Cognitive Evaluation Benchmark of Image Reasoning and Description for Large Vision-Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Song, Xiujie, Wu, Mengyue, Zhu, Kenny Q., Zhang, Chunhao, Chen, Yanyi
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929712176037888
author Song, Xiujie
Wu, Mengyue
Zhu, Kenny Q.
Zhang, Chunhao
Chen, Yanyi
author_facet Song, Xiujie
Wu, Mengyue
Zhu, Kenny Q.
Zhang, Chunhao
Chen, Yanyi
contents Large Vision-Language Models (LVLMs), despite their recent success, are hardly comprehensively tested for their cognitive abilities. Inspired by the prevalent use of the Cookie Theft task in human cognitive tests, we propose a novel evaluation benchmark to evaluate high-level cognitive abilities of LVLMs using images with rich semantics. The benchmark consists of 251 images along with comprehensive annotations. It defines eight reasoning capabilities and comprises an image description task and a visual question answering task. Our evaluation of well-known LVLMs shows that there is still a significant gap in cognitive abilities between LVLMs and humans.
format Preprint
id arxiv_https___arxiv_org_abs_2402_18409
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Cognitive Evaluation Benchmark of Image Reasoning and Description for Large Vision-Language Models
Song, Xiujie
Wu, Mengyue
Zhu, Kenny Q.
Zhang, Chunhao
Chen, Yanyi
Artificial Intelligence
Computation and Language
Computer Vision and Pattern Recognition
Large Vision-Language Models (LVLMs), despite their recent success, are hardly comprehensively tested for their cognitive abilities. Inspired by the prevalent use of the Cookie Theft task in human cognitive tests, we propose a novel evaluation benchmark to evaluate high-level cognitive abilities of LVLMs using images with rich semantics. The benchmark consists of 251 images along with comprehensive annotations. It defines eight reasoning capabilities and comprises an image description task and a visual question answering task. Our evaluation of well-known LVLMs shows that there is still a significant gap in cognitive abilities between LVLMs and humans.
title A Cognitive Evaluation Benchmark of Image Reasoning and Description for Large Vision-Language Models
topic Artificial Intelligence
Computation and Language
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2402.18409