Harness-Zero: Agent-as-Harness Distills Scaffolding into Model Weights
Peking University and Google's Harness-Zero introduces agent-as-harness, where a harnessing agent reviews and corrects a student agent's actions using an evolved reference harness, then distills the corrected trajectories into the student model via SFT, achieving 90% relative improvement on a 9B model without external scaffolding at deployment.
