UniFinEval: Towards Unified Evaluation of Financial Multimodal Models across Text, Images and Videos
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914292832403456 |
|---|---|
| author | Yang, Zhi Zeng, Lingfeng Lou, Fangqi Qi, Qi Zhang, Wei Wu, Zhenyu Yu, Zhenxiong Han, Jun Jin, Zhiheng Zhang, Lejie Huang, Xiaoming Liang, Xiaolong Wei, Zheng Zou, Junbo Cheng, Dongpo Liu, Zhaowei Guo, Xin Zhang, Rongjunchen Zhang, Liwen |
| author_facet | Yang, Zhi Zeng, Lingfeng Lou, Fangqi Qi, Qi Zhang, Wei Wu, Zhenyu Yu, Zhenxiong Han, Jun Jin, Zhiheng Zhang, Lejie Huang, Xiaoming Liang, Xiaolong Wei, Zheng Zou, Junbo Cheng, Dongpo Liu, Zhaowei Guo, Xin Zhang, Rongjunchen Zhang, Liwen |
| contents | Multimodal large language models are playing an increasingly significant role in empowering the financial domain, however, the challenges they face, such as multimodal and high-density information and cross-modal multi-hop reasoning, go beyond the evaluation scope of existing multimodal benchmarks. To address this gap, we propose UniFinEval, the first unified multimodal benchmark designed for high-information-density financial environments, covering text, images, and videos. UniFinEval systematically constructs five core financial scenarios grounded in real-world financial systems: Financial Statement Auditing, Company Fundamental Reasoning, Industry Trend Insights, Financial Risk Sensing, and Asset Allocation Analysis. We manually construct a high-quality dataset consisting of 3,767 question-answer pairs in both chinese and english and systematically evaluate 10 mainstream MLLMs under Zero-Shot and CoT settings. Results show that Gemini-3-pro-preview achieves the best overall performance, yet still exhibits a substantial gap compared to financial experts. Further error analysis reveals systematic deficiencies in current models. UniFinEval aims to provide a systematic assessment of MLLMs' capabilities in fine-grained, high-information-density financial environments, thereby enhancing the robustness of MLLMs applications in real-world financial scenarios. Data and code are available at https://github.com/aifinlab/UniFinEval. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_22162 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | UniFinEval: Towards Unified Evaluation of Financial Multimodal Models across Text, Images and Videos Yang, Zhi Zeng, Lingfeng Lou, Fangqi Qi, Qi Zhang, Wei Wu, Zhenyu Yu, Zhenxiong Han, Jun Jin, Zhiheng Zhang, Lejie Huang, Xiaoming Liang, Xiaolong Wei, Zheng Zou, Junbo Cheng, Dongpo Liu, Zhaowei Guo, Xin Zhang, Rongjunchen Zhang, Liwen General Finance Artificial Intelligence Computation and Language Multimodal large language models are playing an increasingly significant role in empowering the financial domain, however, the challenges they face, such as multimodal and high-density information and cross-modal multi-hop reasoning, go beyond the evaluation scope of existing multimodal benchmarks. To address this gap, we propose UniFinEval, the first unified multimodal benchmark designed for high-information-density financial environments, covering text, images, and videos. UniFinEval systematically constructs five core financial scenarios grounded in real-world financial systems: Financial Statement Auditing, Company Fundamental Reasoning, Industry Trend Insights, Financial Risk Sensing, and Asset Allocation Analysis. We manually construct a high-quality dataset consisting of 3,767 question-answer pairs in both chinese and english and systematically evaluate 10 mainstream MLLMs under Zero-Shot and CoT settings. Results show that Gemini-3-pro-preview achieves the best overall performance, yet still exhibits a substantial gap compared to financial experts. Further error analysis reveals systematic deficiencies in current models. UniFinEval aims to provide a systematic assessment of MLLMs' capabilities in fine-grained, high-information-density financial environments, thereby enhancing the robustness of MLLMs applications in real-world financial scenarios. Data and code are available at https://github.com/aifinlab/UniFinEval. |
| title | UniFinEval: Towards Unified Evaluation of Financial Multimodal Models across Text, Images and Videos |
| topic | General Finance Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2601.22162 |