arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2410.18652cs.LGcs.AIcs.CL
收藏

C²：基于LLM的图表生成的可扩展自动反馈

$C^2$: Scalable Auto-Feedback for LLM-based Chart Generation

Yonsei University（延世大学）；KAIST AI（韩国科学技术院人工智能研究院）

Woosung Koh, Jang Han Yoon, MinHyung Lee, Youngjin Song, Jaegwan Cho, Jaehyun Kang, Taehyeon Kim, Se-Young Yun, Youngjae Yu, Bongshin Lee

备注：
NAACL 2025 Main (Long)
阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对LLM图表生成因数据稀缺、人工整理成本高导致的可扩展性问题，提出无参考自动反馈框架C²，含ChartAF反馈提供器与ChartUIE-8K数据集，实验证明其提升图表生成质量与数据多样性，且94%用户偏好该数据集查询。

摘要：

由于数据有限以及通过人工整理实现扩展的成本高昂，使用大型语言模型（LLMs）生成高质量图表面临重大挑战。〈指令、数据、代码〉三元组稀缺且手动整理成本高昂，因为其创建需要技术专业知识。为解决这一可扩展性挑战，我们引入了一种无参考自动反馈生成器，无需昂贵的人工干预。我们的新型框架C²包含（1）自动反馈提供器（ChartAF，图表自动反馈）和（2）多样化的无参考数据集（ChartUIE-8K）。结果令人信服：在首次实验中，74%的受访者强烈偏好、10%偏好反馈后的结果；第二次反馈后实验表明ChartAF优于9个基线。此外，ChartUIE-8K通过将查询、数据集和图表类型分别比基准提高5982%、1936%和91%，显著提升了数据多样性。最后，针对LLM用户的研究显示，94%的参与者偏好ChartUIE-8K的查询，93%认为其与现实用例对齐。核心贡献已开源，地址为chartsquared.github.io，包含丰富的定性示例。

英文摘要：

Generating high-quality charts with Large Language Models (LLMs) presents significant challenges due to limited data and the high cost of scaling through human curation. $\langle \text{instruction}, \text{data}, \text{code} \rangle$ triplets are scarce and expensive to manually curate as their creation demands technical expertise. To address this scalability challenge, we introduce a reference-free automatic feedback generator, which eliminates the need for costly human intervention. Our novel framework, C$^2$, consists of (1) an automatic feedback provider (ChartAF) and (2) a diverse, reference-free dataset (ChartUIE-8K). The results are compelling: in our first experiment, 74% of respondents strongly preferred, and 10% preferred, the results after feedback. The second post-feedback experiment demonstrates that ChartAF outperform nine baselines. Moreover, ChartUIE-8K significantly improves data diversity by increasing queries, datasets, and chart types by 5982%, 1936%, and 91%, respectively, over benchmarks. Finally, a study of LLM users revealed that 94% of participants preferred ChartUIE-8K's queries, with 93% deeming them aligned with real-world use cases. Core contributions are available as open-source at chartsquared.github.io, with ample qualitative examples.

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。

↑