测试大模型生成误导性图表的漏洞，揭示其安全风险。
ChartAttack: Testing the Vulnerability of LLMs to Malicious Prompting in Chart Generation
- 设计攻击框架ChartAttack，通过注入误导元素生成虚假图表。
- 导致模型问答准确率下降17.2点（域内）和11.9点（跨域）。
- 提供可用来训练模型抗误导能力的数据集AttackViz。
多模态大语言模型（MLLMs）正被广泛用于从数据表自动生成图表，提升分析效率的同时也带来新的滥用风险。我们提出ChartAttack框架，通过在图表设计中注入误导性元素，大规模测试MLLM生成误导性图表的能力。同时构建AttackViz数据集，包含图表规范与问答对，标注了有效误导项及其引发的错误答案。实验显示，ChartAttack使MLLM问答准确率在域内下降17.2点，跨域下降11.9点。受控的人类实验表明，此类误导图表会降低人类答题表现。最后，我们证明AttackViz可用于微调模型以增强对误导图表的鲁棒性。研究凸显了在设计、评估和部署基于MLLM的图表生成系统时亟需加强安全性与鲁棒性考量。代码与数据已公开。
原文摘要 · Abstract (English)
Multimodal large language models (MLLMs) are increasingly used to automate chart generation from data tables, improving analysis and reporting efficiency while introducing new misuse risks. We present ChartAttack, a framework for evaluating how MLLMs can generate misleading charts at scale by injecting misleaders into chart designs to induce incorrect interpretations. We also introduce AttackViz, a chart question-answering (QA) dataset where each (chart specification, QA) pair is labeled with effective misleaders and their induced incorrect answers. ChartAttack significantly degrades QA performance, reducing MLLM accuracy by 17.2 points in-domain and 11.9 cross-domain. A controlled human study shows that misleading charts generated by ChartAttack reduce human chart QA performance. Finally, we demonstrate that AttackViz can be used to fine-tune MLLMs to improve robustness against misleading charts. Our findings highlight an urgent need for robustness and security considerations in the design, evaluation, and deployment of MLLM-based chart generation systems. We make our code and data publicly available.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。