SafetyALFRED：评估多模态大语言模型的安全意识规划
SafetyALFRED: Evaluating Safety-Conscious Planning of Multimodal Large Language Models
AI总结：
本文提出SafetyALFRED，基于ALFRED基准测试，引入六类厨房真实世界危险，评估多模态大语言模型在安全意识规划中的表现，发现静态QA评估不足，需转向强调具身环境中的纠正行动的基准。
摘要：
多模态大语言模型越来越多地被用作交互环境中的自主代理，但其主动解决安全威胁的能力仍不足。我们介绍了SafetyALFRED，基于具身代理基准ALFRED，增加了六类真实厨房危险。尽管现有安全评估侧重于通过脱离身体的问答设置进行危险识别，我们评估了来自Qwen、Gemma和Gemini家族的十一种最新模型，不仅在危险识别方面，还在具身规划中主动进行风险缓解。实验结果揭示了显著的对齐差距：虽然模型在问答设置中能准确识别危险，但这些危险的平均缓解成功率却较低。我们的发现表明，通过问答的静态评估不足以保证物理安全，因此我们倡导向优先考虑具身环境中纠正行动的基准转变。我们通过https://github.com/sled-group/SafetyALFRED.git开源我们的代码和数据集。
英文摘要：
Multimodal Large Language Models are increasingly adopted as autonomous agents in interactive environments, yet their ability to proactively address safety hazards remains insufficient. We introduce SafetyALFRED, built upon the embodied agent benchmark ALFRED, augmented with six categories of real-world kitchen hazards. While existing safety evaluations focus on hazard recognition through disembodied question answering (QA) settings, we evaluate eleven state-of-the-art models from the Qwen, Gemma, and Gemini families on not only hazard recognition, but also active risk mitigation through embodied planning. Our experimental results reveal a significant alignment gap: while models can accurately recognize hazards in QA settings, average mitigation success rates for these hazards are low in comparison. Our findings demonstrate that static evaluations through QA are insufficient for physical safety, thus we advocate for a paradigm shift toward benchmarks that prioritize corrective actions in embodied contexts. We open-source our code and dataset under https://github.com/sled-group/SafetyALFRED.git
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。
