RealChart2Code: Advancing Chart-to-Code Generation with Real Data and Multi-Task Evaluation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915893148123136 |
|---|---|
| author | Zhang, Jiajun Li, Yuying Li, Zhixun Guo, Xingyu Wu, Jingzhuo Zheng, Leqi Yang, Yiran Zhang, Jianke Li, Qingbin Yan, Shannan Li, Zhetong Jia, Changguo Wu, Junfei Wang, Zilei Liu, Qiang Wang, Liang |
| author_facet | Zhang, Jiajun Li, Yuying Li, Zhixun Guo, Xingyu Wu, Jingzhuo Zheng, Leqi Yang, Yiran Zhang, Jianke Li, Qingbin Yan, Shannan Li, Zhetong Jia, Changguo Wu, Junfei Wang, Zilei Liu, Qiang Wang, Liang |
| contents | Vision-Language Models (VLMs) have demonstrated impressive capabilities in code generation across various domains. However, their ability to replicate complex, multi-panel visualizations from real-world data remains largely unassessed. To address this gap, we introduce \textbf{\texttt{RealChart2Code}}, a new large-scale benchmark with over 2,800 instances grounded in authentic datasets and featuring tasks with clear analytical intent. Crucially, it is the first benchmark to systematically evaluate chart generation from large-scale raw data and assess iterative code refinement in a multi-turn conversational setting. Our comprehensive evaluation of 14 leading VLMs on \texttt{RealChart2Code} reveals significant performance degradation compared to simpler benchmarks, highlighting their struggles with complex plot structures and authentic data. Our analysis uncovers a substantial performance gap between proprietary and open-weight models and confirms that even state-of-the-art VLMs often fail to accurately replicate intricate, multi-panel charts. These findings provide valuable insights into the current limitations of VLMs and guide future research directions. We release the benchmark and code at \url{https://github.com/Speakn0w/RealChart2Code}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_25804 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | RealChart2Code: Advancing Chart-to-Code Generation with Real Data and Multi-Task Evaluation Zhang, Jiajun Li, Yuying Li, Zhixun Guo, Xingyu Wu, Jingzhuo Zheng, Leqi Yang, Yiran Zhang, Jianke Li, Qingbin Yan, Shannan Li, Zhetong Jia, Changguo Wu, Junfei Wang, Zilei Liu, Qiang Wang, Liang Computation and Language Vision-Language Models (VLMs) have demonstrated impressive capabilities in code generation across various domains. However, their ability to replicate complex, multi-panel visualizations from real-world data remains largely unassessed. To address this gap, we introduce \textbf{\texttt{RealChart2Code}}, a new large-scale benchmark with over 2,800 instances grounded in authentic datasets and featuring tasks with clear analytical intent. Crucially, it is the first benchmark to systematically evaluate chart generation from large-scale raw data and assess iterative code refinement in a multi-turn conversational setting. Our comprehensive evaluation of 14 leading VLMs on \texttt{RealChart2Code} reveals significant performance degradation compared to simpler benchmarks, highlighting their struggles with complex plot structures and authentic data. Our analysis uncovers a substantial performance gap between proprietary and open-weight models and confirms that even state-of-the-art VLMs often fail to accurately replicate intricate, multi-panel charts. These findings provide valuable insights into the current limitations of VLMs and guide future research directions. We release the benchmark and code at \url{https://github.com/Speakn0w/RealChart2Code}. |
| title | RealChart2Code: Advancing Chart-to-Code Generation with Real Data and Multi-Task Evaluation |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2603.25804 |