arXiv 2601.136062026-04-29cs.CV

ChartVerse: 通过从零开始可靠程序合成实现图表推理的扩展

ChartVerse: Scaling Chart Reasoning via Reliable Programmatic Synthesis from Scratch

Shanghai AI Laboratory（上海人工智能实验室）；Peking University（北京大学）；Shanghai Jiao Tong University（上海交通大学）；Zhongguancun Academy（中关村学院）；Beijing Key Laboratory of Software and Hardware Cooperative Artificial Intelligence Systems（北京市软件与硬件协同人工智能系统重点实验室）

Zheng Liu, Honglin Lin, Chonghan Qin, Xiaoyang Wang, Xin Gao, Yu Li, Mengzhang Cai, Yun Zhu, Zhanping Zhong, Qizhi Pei, Zhuoshi Pan, Xiaoran Shang, Bin Cui, Con… 展开作者

Zheng Liu, Honglin Lin, Chonghan Qin, Xiaoyang Wang, Xin Gao, Yu Li, Mengzhang Cai, Yun Zhu, Zhanping Zhong, Qizhi Pei, Zhuoshi Pan, Xiaoran Shang, Bin Cui, Conghui He, Wentao Zhang, Lijun Wu

备注：: 29 pages

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

ChartVerse通过从零开始生成复杂图表和可靠推理数据，解决开放源模型开发中高质量训练数据缺乏的问题，采用RPE量化图表复杂度并生成高复杂度图表，结合truth-anchored inverse QA合成和CoT蒸馏提升推理深度。

摘要：

图表推理是视觉语言模型（VLMs）的关键能力。然而，开放源模型的发展受到高质量训练数据缺乏的严重阻碍。现有数据集面临双重挑战：合成图表往往过于简单和重复，而相关的问答对容易产生幻觉且缺乏复杂任务所需的推理深度。为弥合这一差距，我们提出了ChartVerse，一个可扩展的框架，旨在从零开始合成复杂图表和可靠的推理数据。（1）为解决简单模式的瓶颈，我们首先引入Rollout Posterior Entropy（RPE），一种新的度量标准，用于量化图表复杂度。受RPE指导，我们开发了复杂度感知的图表编码器，通过可执行程序自主合成多样、高复杂度的图表。（2）为保证推理严谨性，我们开发了truth-anchored inverse QA合成。与标准生成不同，我们采用答案优先范式：从源代码中提取确定性答案，根据这些锚点生成问题，并强制执行严格的 consistency 验证。为进一步提高难度和推理深度，我们根据模型失败率过滤样本，并蒸馏高质量的Chain-of-Thought（CoT）推理。我们使用Qwen3-VL-30B-A3B-Thinking作为教师，精心整理了ChartVerse-SFT-600K和ChartVerse-RL-40K。实验结果表明，ChartVerse-8B实现了最先进的性能，显著超越其教师并与更强大的Qwen3-VL-32B-Thinking相媲美。我们发布我们的代码、模型权重和数据集在https://chartverse.github.io。

英文摘要：

Chart reasoning is a critical capability for Vision Language Models (VLMs). However, the development of open-source models is severely hindered by the lack of high-quality training data. Existing datasets suffer from a dual challenge: synthetic charts are often simplistic and repetitive, while the associated QA pairs are prone to hallucinations and lack the reasoning depth required for complex tasks. To bridge this gap, we propose ChartVerse, a scalable framework designed to synthesize complex charts and reliable reasoning data from scratch. (1) To address the bottleneck of simple patterns, we first introduce Rollout Posterior Entropy (RPE), a novel metric that quantifies chart complexity. Guided by RPE, we develop complexity-aware chart coder to autonomously synthesize diverse, high-complexity charts via executable programs. (2) To guarantee reasoning rigor, we develop truth-anchored inverse QA synthesis. Diverging from standard generation, we adopt an answer-first paradigm: we extract deterministic answers directly from the source code, generate questions conditional on these anchors, and enforce strict consistency verification. To further elevate difficulty and reasoning depth, we filter samples based on model fail-rate and distill high-quality Chain-of-Thought (CoT) reasoning. We curate ChartVerse-SFT-600K and ChartVerse-RL-40K using Qwen3-VL-30B-A3B-Thinking as the teacher. Experimental results demonstrate that ChartVerse-8B achieves state-of-the-art performance, notably surpassing its teacher and rivaling the stronger Qwen3-VL-32B-Thinking. We release our code, model weights, and datasets in https://chartverse.github.io.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。