Effectiveness assessment of recent large vision-language models
Visual Intelligence · 28.06.2024 · 2 (1)
DOI: 10.1007/s44267-024-00050-1 ↗Kurz gefasst
Fachartikel in der Zeitschrift Visual Intelligence, online erschienen am 28.06.2024. Verfasst von Yao Jiang und 7 weiteren Autor:innen. Die Arbeit wurde bisher 45-mal zitiert (Crossref) und nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Fundamental Research Funds for the Central Universities.
Abstract (Originalsprache)
The advent of large vision-language models (LVLMs) represents a remarkable advance in the quest for artificial general intelligence. However, the models’ effectiveness in both specialized and general tasks warrants further investigation. This paper endeavors to evaluate the competency of popular LVLMs in specialized and general tasks, respectively, aiming to offer a comprehensive understanding of these novel models. To gauge their effectiveness in specialized tasks, we employ six challenging tasks in three different application scenarios: natural, healthcare, and industrial. These six tasks include salient/camouflaged/transparent object detection, as well as polyp detection, skin lesion detection, and industrial anomaly detection. We examine the performance of three recent open-source LVLMs, including MiniGPT-v2, LLaVA-1.5, and Shikra, on both visual recognition and localization in these tasks. Moreover, we conduct empirical investigations utilizing the aforementioned LVLMs together with GPT-4V, assessing their multi-modal understanding capabilities in general tasks including object counting, absurd question answering, affordance reasoning, attribute recognition, and spatial relation reasoning. Our investigations reveal that these LVLMs demonstrate limited proficiency not only in specialized tasks but also in general tasks. We delve deep into this inadequacy and uncover several potential factors, including limited cognition in specialized tasks, object hallucination, text-to-image interference, and decreased robustness in complex problems. We hope that this study can provide useful insights for the future development of LVLMs, helping researchers improve LVLMs for both general and specialized applications.
Förderung
- National Natural Science Foundation of China
- Fundamental Research Funds for the Central Universities
Zitieren
APA
Jiang, Y., Yan, X., Ji, G., Fu, K., Sun, M., Xiong, H., Fan, D., & Khan, F. S. (2024). Effectiveness assessment of recent large vision-language models. Visual Intelligence, 2(1). https://doi.org/10.1007/s44267-024-00050-1
Vancouver
Jiang Y, Yan X, Ji G, Fu K, Sun M, Xiong H, et al. Effectiveness assessment of recent large vision-language models. Visual Intelligence. 2024;2(1). doi:10.1007/s44267-024-00050-1
BibTeX
@article{Jiang2024,
author = {Jiang, Yao and Yan, Xinyu and Ji, Ge-Peng and Fu, Keren and Sun, Meijun and Xiong, Huan and Fan, Deng-Ping and Khan, Fahad Shahbaz},
title = {Effectiveness assessment of recent large vision-language models},
journal = {Visual Intelligence},
year = {2024},
volume = {2},
number = {1},
doi = {10.1007/s44267-024-00050-1}
}
EUVIMED