safety security
A Smaller Agent Decides When to Call the Expensive One
LEASH uses latent uncertainty for routing, step-level escalation and a shadow observer after a strong black-box agent takes over.
Summary
LEASH uses latent uncertainty for routing, step-level escalation and a shadow observer after a strong black-box agent takes over.
Using a frontier model at every computer-use step is costly, but leaving the whole task to a smaller model often fails. LEASH compares a weak model's hidden state with its past successful and failed trajectories, escalating when the state resembles failure. After handoff, the weak model continues as a shadow observer because the stronger API exposes no internals. On OSWorld across four backbones, the authors report gains of 6.55 to 11.94 points over random escalation at matched rates. One configuration reached 73.66 percent task success at $0.4117 per task, roughly matching the paper's always-on Claude Opus 5 result while reducing inference cost 19.2 percent. These figures are benchmark- and pricing-specific, not a deployment guarantee.
Why it matters
LEASH uses latent uncertainty for routing, step-level escalation and a shadow observer after a strong black-box agent takes over.
Limits and context
- These figures are benchmark- and pricing-specific, not a deployment guarantee.
Key claims
LEASH uses latent uncertainty for routing, step-level escalation and a shadow observer after a strong black-box agent takes over.
Qualification: These figures are benchmark- and pricing-specific, not a deployment guarantee.
Evidence: source-2026-10-16-010
Sources
- arXiv preprint 2610.22280, original v1arXiv / submitting research authors · primary research
Corrections
No corrections have been recorded for this story.