发表机构
William & Mary; IBM; Allen Institute for AI; Oak Ridge National Laboratory; Northwestern University（威廉玛丽学院; 国际商业机器公司; 艾伦人工智能研究所; 橡树岭国家实验室; 西北大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对图表、表格、代码跨表示学习的挑战，提出CoCoEvolve方法，通过定义一一对应关系、利用表示一致性优化模型，在4个基准的训练和测试场景中提升了性能。
AI 中文摘要
随着图表图像、表格数据和可视化代码在不同领域发挥日益重要的作用，这些模态间的跨表示理解对AI系统构成了根本性挑战：表示间的关系本质上是一对多的，监督信号模糊且成本高昂，模型优化缺乏既具有方向适应性又能泛化到任务特定目标之外的通用信号。我们提出CoCoEvolve以提升图表、表格和代码表示间的一致性。我们未将跨表示映射视为一对多问题，而是定义显式的一一对应关系，并利用表示间的一致性优化模型，无需额外标注。训练期间，CoCoEvolve@Train在图表-表格-代码循环中执行协同演化；CoCoEvolve@Test在推理时应用相同的一致性目标以进行测试时协同优化。我们还推出CoCoEvolve@Eval，这是一套涵盖全部6项跨表示任务的评估套件。在4个基准测试中，CoCoEvolve在训练时和测试时设置下均提升了性能。项目页面：this https URL。
英文摘要
As chart images, tabular data, and visualization code play increasingly important roles across diverse domains, cross-representation understanding across these modalities poses fundamental challenges for AI systems: the relationships across representations are inherently \textit{one-to-many}, supervision is ambiguous and costly, and model optimization lacks a principled signal that is both direction-adaptive and representation-generalizable beyond task-specific objectives. We introduce CoCoEvolve to improve consistency across chart, table, and code representations. Instead of treating cross-representation mapping as a one-to-many problem, we define explicit one-to-one correspondences and optimize models using agreement between representations, without additional annotations. During training, CoCoEvolve@Train performs co-evolution across the chart-table-code cycle, while CoCoEvolve@Test applies the same consistency objective at inference time for test-time co-optimization. We also present CoCoEvolve@Eval, an evaluation suite covering all six cross-representation tasks. Across four benchmarks, CoCoEvolve improves performance in both training-time and test-time settings. Our project page: https://xhguo7.github.io/CoCoEvolve/.
