arXiv cs.CL· Zheyuan Yang, Liqiang Shang, Junjie Chen, Xun Yang, Chenglong Xu, Bo Yuan, Chenyuan Jiao, Yaoru Sun, Yilun Zhao·· 9 小时前AI 评分31
TableVista:多模态表格推理在视觉与结构复杂度下的基准测试
TableVista: Benchmarking Multimodal Table Reasoning under Visual and Structural Complexity
AI 导读
TableVista 是一个评估基础模型多模态表格推理能力的基准,包含 3,000 道表格推理题,每题经多风格渲染与变换扩展为 10 个视觉变体,共 30,000 个多模态样本。研究团队在 TableVista 上评测了 29 个开源与闭源前沿模型,发现模型在不同渲染风格下表现基本稳定,但在复杂结构布局和纯视觉设置下性能明显下降。
正文 · 原文
Abstract:We introduce TableVista, a comprehensive benchmark for evaluating foundation models in multimodal table reasoning under visual and structural complexity. TableVista consists of 3,000 high-quality table reasoning problems, where each instance is expanded into 10 distinct visual variants through our multi-style rendering and transformation pipeline. This process encompasses diverse scenario styles, robustness perturbations, and vision-only configurations, culminating in 30,000 multimodal samples for a multi-dimensional evaluation. We conduct an extensive evaluation of 29 state-of-the-art open-source and proprietary foundation models on TableVista. Through comprehensive quantitative and qualitative analysis, we find that while evaluated models remain largely stable across diverse rendering styles, they exhibit pronounced performance degradation on complex structural layouts and vision-only settings, revealing that current models struggle to maintain reasoning consistency when structural complexity combines with visually integrated presentations. These findings highlight critical gaps in current multimodal capabilities, providing insights for advancing more robust and reliable table understanding models.
| Comments: | ACL 2026 Findings |
| Subjects: | Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2605.05955 [cs.CL] |
| (or arXiv:2605.05955v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2605.05955 arXiv-issued DOI via DataCite |
Submission history
From: Zheyuan Yang [view email]
[v1]
Thu, 7 May 2026 10:03:25 UTC (1,631 KB)
[v2]
Tue, 17 Nov 2026 12:16:26 UTC (8,961 KB)
来源:arXiv cs.CL · arxiv.org