Concept detail
on-policy RL (GRPO)
0 tagged videos · 1 connection videos · degree 0
Publication timeline
- 2026-091 video
Source videos
Observed neighbors
reward model
Rollouts sampled from the current policy are scored by a reward model, with a KL penalty anchoring the policy to its reference.
1 source videos
AI Safety (CS 2881) Fall 2026 Lecture 3: Modern LLM Training and InferenceRelated tagged insights
No core insights are tagged with this concept.