国际讯息

Forecasting Future Behavior as a Learning Task

来源：arXiv AI 作者：Mosh Levy, Yoav Goldberg, Asa Cooper Stickland 阅读原文 →
cs.AI

arXiv:2606.11445v1 Announce Type: new Abstract: Trust in an AI system is often anchored by explanations of how it works, which one then uses to forecast its behavior on new inputs. For large reasoning models (LRMs), this conventional route is particularly difficult to follow: explanation methods...