existing methods typically attempt to address both issues simultaneously using a single paradigm
大多数人认为解决长时程LLM代理问题应该采用统一的方法同时处理全局进度和局部可行性，但作者认为这两种挑战本质上是不同的：一个依赖模糊语义规划，另一个需要严格逻辑约束和状态验证。这种分离的观点挑战了当前AI研究的主流范式。
2 Matching Annotations
- Last 7 days
-
arxiv.org arxiv.org
Tags
Annotators
URL
-
- Dec 2021
-
onlinelibrary.wiley.com onlinelibrary.wiley.com
-
Sloman, S. A. (2021). How Do We Believe? Topics in Cognitive Science, 0(2021), 1–14. https://doi.org/10.1111/tops.12580
Tags
- human thought
- pattern recognition
- unfamiliar circumstance
- sophisticated associative model
- representational scheme
- generalizability
- information processing
- memory
- causal reasoning
- lang:en
- knowledge
- cognitive science
- is:article
- representational language
- dual system of thinking
- predictability
Annotators
URL
-