SPA-VL：面向视觉语言模型的综合安全偏好对齐数据集
SPA-VL: A Comprehensive Safety Preference Alignment Dataset for Vision Language Model
AI总结：
针对视觉语言模型安全对齐研究不足、缺乏高质量大规模数据集的问题，提出覆盖多类有害场景、含十万余样本的自动化构建偏好数据集SPA-VL，可有效提升模型对齐后的无害性与有用性。
摘要：
视觉语言模型（VLM）的出现为多模态信息理解带来了前所未有的进展。VLM中文本与视觉语义的结合高度复杂且多样，使得这类模型的安全对齐颇具挑战。此外，由于针对VLM安全对齐的研究有限，目前缺乏大规模、高质量的数据集。为解决这些局限，我们提出了一个面向视觉语言模型的安全偏好对齐数据集，命名为SPA-VL。在广度上，SPA-VL覆盖6个有害性领域、13个类别和53个子类别，包含100788个由（问题、图像、优选响应、拒选响应）组成的四元组样本。在深度上，这些响应收集自12个开源（如QwenVL）和闭源（如Gemini）VLM，以确保多样性。偏好数据的构建是完全自动化的，实验结果表明，在SPA-VL数据集上使用对齐技术训练的模型，在保持核心能力的同时，无害性和有用性均得到显著提升。SPA-VL作为一个大规模、高质量且多样化的数据集，是确保VLM同时实现无害性和有用性的重要里程碑。
英文摘要：
The emergence of Vision Language Models (VLMs) has brought unprecedented advances in understanding multimodal information. The combination of textual and visual semantics in VLMs is highly complex and diverse, making the safety alignment of these models challenging. Furthermore, due to the limited study on the safety alignment of VLMs, there is a lack of large-scale, high-quality datasets. To address these limitations, we propose a Safety Preference Alignment dataset for Vision Language Models named SPA-VL. In terms of breadth, SPA-VL covers 6 harmfulness domains, 13 categories, and 53 subcategories, and contains 100,788 samples of the quadruple (question, image, chosen response, rejected response). In terms of depth, the responses are collected from 12 open-source (e.g., QwenVL) and closed-source (e.g., Gemini) VLMs to ensure diversity. The construction of preference data is fully automated, and the experimental results indicate that models trained with alignment techniques on the SPA-VL dataset exhibit substantial improvements in harmlessness and helpfulness while maintaining core capabilities. SPA-VL, as a large-scale, high-quality, and diverse dataset, represents a significant milestone in ensuring that VLMs achieve both harmlessness and helpfulness.
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。
