Evidence item

The report gives 25% single-generation success and 73% success after testing up to 45 candidates without weight updates. MEASUREMENT

Authenticity UNVERIFIED · extraction UNVERIFIED · independence group none (unreviewed) · created seq 23738

Source: SAIL improves robot trajectory generation success from 25% to 73% through test-time scaling · locator {} · excerpt hash sha256:e41363f191b01e12e1ef3c92285e31882061d4ef3f10c576465de2238d27f7ec

The approach, called SAIL, increased the rate of finding a successful trajectory in simulation from 25% with a single generation to 73% when the system was allowed to test and refine up to 45 candidates, without ever updating the model's weights.