arXiv 2601.096132026-01-14cs.CVcs.AI

CogRail: 对智能铁路运输系统中认知入侵感知的视觉语言模型进行基准测试

CogRail: Benchmarking VLMs in Cognitive Intrusion Perception for Intelligent Railway Transportation Systems

State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences（多模态人工智能系统国家重点实验室，自动化研究所，中国科学院）；School of Automation, Beijing Institute of Technology（自动化学院，北京理工大学）；Signal & Communication Research Institute, China Academy of Railway Sciences（信号与通信研究所，中国铁路科学研究院）；Beijing Huairou Academy of Parallel Sensing（北京怀柔平行感知院）；School of Computer Science and Technology, Beijing Jiaotong University（计算机科学与技术学院，北京交通大学）；State Key Lab of Intelligent Transportation Systems, School of Transportation Science and Engineering, Beihang University（智能交通系统国家重点实验室，交通科学与工程学院，北京航空航天大学）

Yonglin Tian, Qiyao Zhang, Wei Xu, Yutong Wang, Yihao Wu, Xinyi Li, Xingyuan Dai, Hui Zhang, Zhiyong Cui, Baoqing Guo, Zujun Yu, Yisheng Lv

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

CogRail提出一个用于评估视觉语言模型在认知入侵感知任务中性能的基准，通过联合微调框架提升模型在时空推理和威胁分析中的表现。

摘要：

准确且早期感知潜在入侵目标对于确保铁路运输系统安全至关重要。然而，大多数现有系统仅限于在固定视觉范围内进行物体分类，并使用基于规则的启发法来确定入侵状态，往往忽视那些可能造成潜在入侵风险的目标。预见此类风险需要对感兴趣对象（OOI）的空间上下文和时间动态进行认知，这对传统视觉模型提出了挑战。为了促进深入的入侵感知，我们引入了一个新的基准，CogRail，它整合了经过筛选的开源数据集和具有认知驱动的问答注释，以支持时空推理和预测。基于此基准，我们系统地评估了最先进的视觉语言模型（VLMs），使用多模态提示来识别它们在此领域的优缺点。此外，我们对VLMs进行微调以提高性能，并提出一个联合微调框架，整合三个核心任务：位置感知、运动预测和威胁分析，从而促进通用基础模型向专门针对认知入侵感知的专用模型的有效适应。广泛的实验表明，当前大规模多模态模型在需要认知入侵感知任务的复杂时空推理方面存在困难，凸显了现有基础模型在此安全关键领域中的局限性。相比之下，我们提出的联合微调框架通过使模型能够针对领域特定的推理需求进行定向适应，显著提高了模型性能，突显了结构化多任务学习在提高准确性和可解释性方面的优势。代码将在https://github.com/Hub-Tian/CogRail上提供。

英文摘要：

Accurate and early perception of potential intrusion targets is essential for ensuring the safety of railway transportation systems. However, most existing systems focus narrowly on object classification within fixed visual scopes and apply rule-based heuristics to determine intrusion status, often overlooking targets that pose latent intrusion risks. Anticipating such risks requires the cognition of spatial context and temporal dynamics for the object of interest (OOI), which presents challenges for conventional visual models. To facilitate deep intrusion perception, we introduce a novel benchmark, CogRail, which integrates curated open-source datasets with cognitively driven question-answer annotations to support spatio-temporal reasoning and prediction. Building upon this benchmark, we conduct a systematic evaluation of state-of-the-art visual-language models (VLMs) using multimodal prompts to identify their strengths and limitations in this domain. Furthermore, we fine-tune VLMs for better performance and propose a joint fine-tuning framework that integrates three core tasks, position perception, movement prediction, and threat analysis, facilitating effective adaptation of general-purpose foundation models into specialized models tailored for cognitive intrusion perception. Extensive experiments reveal that current large-scale multimodal models struggle with the complex spatial-temporal reasoning required by the cognitive intrusion perception task, underscoring the limitations of existing foundation models in this safety-critical domain. In contrast, our proposed joint fine-tuning framework significantly enhances model performance by enabling targeted adaptation to domain-specific reasoning demands, highlighting the advantages of structured multi-task learning in improving both accuracy and interpretability. Code will be available at https://github.com/Hub-Tian/CogRail.

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。