SOURCE-LINKED INTELLIGENCE

MInTRL: Off-policy Intervention can boost On-policy RL

arXiv · AI, language, vision and robotics · article · Sep 11, 2026 · UTC

Reinforcement learning with verifiable rewards is typically performed on-policy, keeping training data close to the current policy but limiting learning to trajectories that the policy can discover itself. Off-policy methods such as supervised fine-tuning, on the other hand, can leverage external knowledge beyond the base model's capabilities, but may suffer from large distribution shift. The key challenge is thus to expand exploration without sacrificing learnability. In this work, we introduce Minimal Intervention Reinforcement Learning (MInTRL), which expands the exploration frontier throug

Read original source ↗ Open in workspace

recordType

paper

region

Global