自主地球观测智能体的行为自我监控：属性可靠性监控器（ARM）
Behavioural self-monitoring for autonomous earth observation agents: The attribute reliability monitor (ARM)
AI总结：
本研究提出属性可靠性监控器（ARM）框架，通过黑盒行为监控评估语义属性可靠性，可适配不同检测器与监控方法，经卫星热红外图像专用视觉语言模型验证，能区分操作怀疑与语义正确性，指导模型输出使用。
摘要：
语言模型与视觉语言模型可为自主系统提供支持，但运行条件的变化可能会损害其语义输出的可靠性。本研究提出属性可靠性监控器（ARM）这一框架，该框架会随时间累积怀疑证据以评估单个语义属性。ARM可适配不同的检测器与时间监控方法，同时将干预策略交由使用该框架的系统决定。研究人员采用专为卫星热红外图像开发的专用视觉语言模型对ARM的初始实现进行评估，监控采用黑盒方法，仅使用可观测的响应，不访问模型内部状态、置信度分数或外部参考答案。在受控输入退化条件下，响应间的不一致为选择性暂停属性提供了证据。结果表明，外部行为监控可指导模型输出的持续使用，同时区分操作怀疑与语义正确性。
英文摘要：
Language and vision language models can support autonomous systems, but changing operating conditions may compromise the reliability of their semantic outputs. This work introduces the Attribute Reliability Monitor (ARM), a framework that accumulates evidence of suspicion over time to assess individual semantic attributes. ARM accommodates different detectors and temporal monitoring methods, while leaving intervention policies to the consuming system. An initial implementation was evaluated using a specialized vision language model developed for satellite thermal infrared imagery. Monitoring followed a black box approach, using only observable responses without access to internal model states, confidence scores, or external reference answers. Under controlled input degradation, disagreement between responses provided evidence for selective suspension of attributes. The results illustrate how external behavioral monitoring can inform continued use of model outputs, while distinguishing operational suspicion from semantic correctness.
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。
