用嵌入向量距离定义智能体行为规范，让系统按预期方式感知世界。
Pretrained Embeddings as a Behavior Specification Mechanism
- 将现实概念的嵌入向量作为规范语言的一等公民，用理想与观测向量间的距离描述行为属性。
- 提出嵌入时序逻辑（ETL），可表达比以往更广泛的AI系统行为特性。
- 在基于大模型的机器人规划任务中验证有效，能引导系统趋向期望行为。
我们提出一种形式化方法，用于指定依赖感知模型与物理世界交互的系统的行为属性。核心思想是将嵌入——现实概念的数学表示——作为规范语言中的第一类构造，行为属性通过一对理想与观测嵌入之间的距离来表达。为实现该方法，我们提出一种新型时序逻辑——嵌入时序逻辑（ETL），并说明其如何比以往更广泛地描述人工智能系统的属性。通过初步评估，在由基础模型驱动的机器人规划任务中展示了ETL的适用性，结果表明基于嵌入的规范可有效引导系统朝向期望行为演进。
原文摘要 · Abstract (English)
We propose an approach to formally specifying the behavioral properties of systems that rely on a perception model for interactions with the physical world. The key idea is to introduce embeddings -- mathematical representations of a real-world concept -- as a first-class construct in a specification language, where properties are expressed in terms of distances between a pair of ideal and observed embeddings. To realize this approach, we propose a new type of temporal logic called Embedding Temporal Logic (ETL), and describe how it can be used to express a wider range of properties about AI-enabled systems than previously possible. We demonstrate the applicability of ETL through a preliminary evaluation involving planning tasks in robots that are driven by foundation models; the results are promising, showing that embedding-based specifications can be used to steer a system towards desirable behaviors.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。