Visual Question-Answering (VQA) has become key to user experience, particularly after improved generalization capabilities of Vision-Language Models (VLMs). But evaluating VLMs for an application requirement using a standardized framework in practical settings is still challenging. This paper aims to solve that using an end-to-end framework. We present VQA360 - a novel dataset derived from established VQA benchmarks, annotated with task types, application domains, and knowledge types, for a comprehensive evaluation. We also introduce GoEval, a multimodal evaluation metric developed using GPT-4o, achieving a correlation factor of 56.71% with human judgments. Our experiments with state-of-the-art VLMs reveal that no single model excels universally, thus, making a right choice a key design decision. Proprietary models such as Gemini-1.5-Pro and GPT-4o-mini generally outperform others, but open-source models like InternVL-2-8B and CogVLM-2-Llama-3-19B also demonstrate competitive strengths, while providing additional advantages. Our framework can also be extended to other tasks.
@article{arxiv.2409.09269,
title = {Guiding Vision-Language Model Selection for Visual Question-Answering Across Tasks, Domains, and Knowledge Types},
author = {Neelabh Sinha and Vinija Jain and Aman Chadha},
journal= {arXiv preprint arXiv:2409.09269},
year = {2024}
}
Comments
Accepted at The First Workshop of Evaluation of Multi-Modal Generation (EvalMG) in 31st International Conference on Computational Linguistics (COLING), 2025. 8 pages + references + 6 pages of Appendix