Search papers, labs, and topics across Lattice.
The paper introduces ChartVerse, a framework for generating complex synthetic charts and reliable question-answering pairs to train vision-language models (VLMs) for chart reasoning. They address the limitations of existing datasets by using Rollout Posterior Entropy (RPE) to guide the synthesis of diverse, high-complexity charts and employing truth-anchored inverse QA synthesis to ensure reasoning rigor. Experiments show that a ChartVerse-trained 8B model achieves state-of-the-art performance, outperforming its teacher model and rivaling a larger model.
Forget simplistic synthetic data: ChartVerse generates complex charts and reliable reasoning data from scratch, enabling an 8B model to outperform its 30B teacher in chart reasoning.
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.
