arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2606.09428cs.CL
收藏

引导我出去:危机场景下评估VLM操作员通信的框架

Guide Me Out: A Framework to Benchmark VLM Operators Communication in Crisis Scenarios

Fondazione Bruno Kessler(布鲁诺·凯斯勒基金会);University of Trento(特伦托大学)

Giacomo Gonella, Stefano Menini, Marco Guerini

AI总结:

提出一个基准框架,评估视觉语言模型在模拟疏散中引导平民的策略(窄播 vs. 广播)、环境表示(视觉 vs. 图)和威胁行为(静态 vs. 移动),发现窄播降低失败率,视觉表示主导性能,移动威胁增加失败率。

摘要:

有效的危机响应需要空间定位的通信,将平民的语言指导与物理环境联系起来,考虑结构瓶颈、不断变化的威胁和代理特定背景。然而,当前危机通信中的NLP研究主要局限于静态、纯文本分类设置,忽视了AI操作员在动态、具身场景中的关键通信作用。我们通过一个新的基准框架来解决这一差距,该框架用于评估视觉语言模型(VLM)在模拟疏散中引导平民代理的任务。我们测试了两种通信策略(窄播与广播)、两种环境表示(视觉与基于图)和两种威胁行为(静态与移动),跨越九张不同结构复杂度的地图。我们的结果表明,与广播相比,窄播在所有难度级别上持续降低平民失败率。指导质量很大程度上取决于VLM操作员如何表示世界:视觉模态驱动性能,而添加邻接图则依赖于模型且通常有害。移动威胁在所有条件下提高失败率,因为通信必须随时间持续适应。这些发现共同表明,将VLM作为AI操作员部署在疏散场景中仍然是一个非平凡挑战,其中通信策略和输入表示的选择可以直接决定干预的成功或失败。

英文摘要:

Effective crisis response requires spatially grounded communication that bridges linguistic guidance of civilians with the physical environment, accounting for structural bottlenecks, evolving threats, and agent-specific contexts. Yet, current NLP research in crisis communication remains mainly limited to static, text-only classification settings, overlooking the critical communicative role of AI operators in dynamic, embodied scenarios. We address this gap with a novel benchmarking framework for evaluating Vision-Language Models (VLMs) tasked with guiding civilian agents through simulated evacuations. We test two communication strategies (narrowcast vs. broadcast), two environment representations (visual vs. graph-based), and two threat behaviors (static vs. moving) across nine maps of varying structural complexity. Our results show that Narrowcast consistently reduces civilian Fail rates compared to Broadcast across all difficulty levels. Guidance quality depends heavily on how the VLM operator represents the world: the visual modality drives performance, while adding an adjacency graph is model-dependent and often harmful. Moving threats raise Fail rates across all conditions as communication must continuously adapt over time. Together, these findings show that deploying VLMs as AI operators in evacuation scenarios remains a non-trivial challenge, where the choice of communication strategy and input representation can directly determine the success or failure of the intervention.

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。

↑