LEASH: Uncertainty-Aware Weak-to-Strong Escalation and Oversight for Computer-Use Agents

Key points

• Computer-use agents increasingly tackle long-horizon tasks, where running a frontier model at every step is expensive and a small open-weight model alone often fails.1
• Weak-to-strong escalation offers a middle ground where a weak agent acts by default and hands control to a strong one only when needed.1
• LEASH has three layers: (1) initial task uncertainty for early routing, (2) step-level uncertainty for runtime escalation, and, after handoff, (3) a shadow observer for continued oversight of the strong black-box agent.1
• Experiments on OSWorld across four contemporary model backbones show that LEASH consistently improves over standalone weak agents and outperforms random escalation by 6.55--11.94 points at matched escalation rates.1

Sources (2)

• [1]LEASH: Uncertainty-Aware Weak-to-Strong Escalation and Oversight for Computer-Use Agents
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 15, 05:40 PM

  To this end, we propose LEASH, a training-free framework that unleashes the strong agent only when the weak agent's latent uncertainty signals it is needed, and keeps it on a leash afterward.

  Computer-use agents increasingly tackle long-horizon tasks, where running a frontier model at every step is expensive and a small open-weight model alone often fails.

• [2]LEASH: Uncertainty-Aware Weak-to-Strong Escalation and Oversight for Computer-Use Agents
  Hugging Face Daily Papers · Oct 15, 12:00 AM · same content

Extractive summary: sentences quoted from the sources.