arXiv cs.AI· Qi Wang, Peijie Wang, Jun Cao, Haoyang Tong, Xiao-Hui Li, Fei Yin, Cheng-Lin Liu·· 12 小时前AI 评分38
ChartEvo：面向长尾图表推理的自进化智能体框架
ChartEvo: Self-Evolving Agent Harnesses for Long-Tail Chart Reasoning
AI 导读
ChartEvo 是一个递归自我改进框架，让图表智能体通过经验驱动的框架级自我进化适应表现不佳的图表类型，并联合进化可复用的图表技能与可执行工具。在 ChartBench 和 ChartX 的长尾图表上，ChartEvo 相比基座模型 Qwen3.7-Plus 分别提升准确率 45.6% 和 32.5%。该框架还学习可复用的局部工具调用链，以组合进化后的能力并控制工具增强推理的推理开销。
正文 · 原文
Abstract:Multimodal large language models (MLLMs) have substantially advanced chart understanding and reasoning, yet their performance remains highly uneven across chart types. While modern MLLMs perform well on common charts, their accuracy drops sharply on long-tail charts with diverse visual structures and reasoning requirements. Tool-augmented agents can improve complex chart reasoning by introducing specialized visual capabilities, but manually engineering strategies and tools for the heterogeneous and evolving long tail is difficult to scale. We introduce \textbf{ChartEvo}, a recursive self-improvement framework that enables chart agents to adapt to underperforming chart types through experience-driven harness-level self-improvement. ChartEvo jointly evolves reusable chart-specific Skills and executable Tools, while learning reusable local tool-call chains to compose evolved capabilities and control the inference overhead of tool-augmented reasoning. On long-tail charts from ChartBench and ChartX, ChartEvo improves accuracy over its base MLLM, Qwen3.7-Plus, by \textbf{45.6\%} and \textbf{32.5\%}, respectively, demonstrating the effectiveness of recursive self-improvement for long-tail chart adaptation.
| Subjects: | Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2610.28282 [cs.AI] |
| (or arXiv:2610.28282v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.28282 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Qi Wang [view email]
[v1]
Tue, 20 Oct 2026 14:34:54 UTC (3,619 KB)
来源：arXiv cs.AI · arxiv.org