VT-Bench: A Unified Benchmark for Visual-Tabular Multi-Modal Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910260067827712 |
|---|---|
| author | Jia, Zi-Yi Cheng, Zi-Jian Zhang, Xin-Yue Yu, Kun-Yang Zhou, Zhi Li, Yu-Feng Guo, Lan-Zhe |
| author_facet | Jia, Zi-Yi Cheng, Zi-Jian Zhang, Xin-Yue Yu, Kun-Yang Zhou, Zhi Li, Yu-Feng Guo, Lan-Zhe |
| contents | Multi-model learning has attracted great attention in visual-text tasks. However, visual-tabular data, which plays a pivotal role in high-stakes domains like healthcare and industry, remains underexplored. In this paper, we introduce \textit{VT-Bench}, the first unified benchmark for standardizing vision-tabular discriminative prediction and generative reasoning tasks. VT-Bench aggregates 14 datasets across 9 domains (medical-centric, while covering pets, media, and transportation) with over 756K samples. We evaluate 23 representative models, including unimodal experts, specialized visual-tabular models, general-purpose vision-language models (VLMs), and tool-augmented methods, highlighting substantial challenges of visual-tabular learning. We believe VT-Bench will stimulate the community to build more powerful multi-modal vision-tabular foundation models.
Benchmark: https://github.com/Ziyi-Jia990/VT-Bench |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_08146 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | VT-Bench: A Unified Benchmark for Visual-Tabular Multi-Modal Learning Jia, Zi-Yi Cheng, Zi-Jian Zhang, Xin-Yue Yu, Kun-Yang Zhou, Zhi Li, Yu-Feng Guo, Lan-Zhe Computer Vision and Pattern Recognition Artificial Intelligence Multi-model learning has attracted great attention in visual-text tasks. However, visual-tabular data, which plays a pivotal role in high-stakes domains like healthcare and industry, remains underexplored. In this paper, we introduce \textit{VT-Bench}, the first unified benchmark for standardizing vision-tabular discriminative prediction and generative reasoning tasks. VT-Bench aggregates 14 datasets across 9 domains (medical-centric, while covering pets, media, and transportation) with over 756K samples. We evaluate 23 representative models, including unimodal experts, specialized visual-tabular models, general-purpose vision-language models (VLMs), and tool-augmented methods, highlighting substantial challenges of visual-tabular learning. We believe VT-Bench will stimulate the community to build more powerful multi-modal vision-tabular foundation models. Benchmark: https://github.com/Ziyi-Jia990/VT-Bench |
| title | VT-Bench: A Unified Benchmark for Visual-Tabular Multi-Modal Learning |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2605.08146 |