Ruslan Salakhutdinov

Faster computer-use agents are bottlenecked by their speed, but existing benchmarks are irreproducible; this new standardized framework reveals counterintuitive speed-efficiency trade-offs and that current open-weight models lag behind.

Long-horizon agents plateau under naive inference scaling unless compute is explicitly redirected from task workers to a dedicated meta-reasoning controller that manages execution flow.

Modifying executable agent harnesses rather than model weights allows test-time adaptation to generalize across unseen reasoning tasks without a single gradient step.