single mention Β· post-training
Group-relative length penalty
A group-relative length penalty is a reinforcement-learning reward adjustment that deducts reward from successful rollouts longer than a reference length derived from the prompt's group of sampled rollouts.1 The deduction applies only to successful rollouts, and a minimum pass-rate gate skips difficult groups to preserve room for exploration while curbing excessive token growth.2
How this definition was written
Summarised by glm-5.3-flash from 1 source; every sentence is checked against the quote it cites.
we compute the reference length ββ π = Quantileπ΅/100{βπ : π β P π} and obtain the adjusted reward
β MiMo V2 6 technical reportThe indicator1[Β·] restricts the deduction to successful rollouts
β MiMo V2 6 technical report
Filed under post-training βΊ reinforcement learning algorithm