offline-training

Introduces Q-Steer, a rollout-time action-value guidance for molecular policy optimization that uses a frozen prefix-action value scorer to improve sampling without changing the online oracle budget.

tldraw announces that neural nets can now be trained offline within the drawing tool.

This paper introduces Feedback Manipulation Regularization (FMR), an algorithm-agnostic method that uses evaluative feedback to improve alignment in imitation learning, achieving up to 98% reduction in misalignment in Safety Gymnasium environments.