Yuzhe Yang
Titles as the sources gave them. In 1 entry, with 0 checked quotations.
In the Base
Every entry they appear in.
LEASH: Uncertainty-Aware Weak-to-Strong Escalation and Oversight for Computer-Use Agents
According to the abstract excerpt, the authors propose LEASH, a training-free framework in which a small agent acts by default and hands control to a stronger one only when its uncertainty signals need. A shadow observer then oversees the strong black-box agent after handoff.