Standard GRPO at 1M scale: why no critic models, and what counts as "behaviors"?
stochasticchasm · x · 2026-09-22
stochasticchasm analyzes a RL training setup run at 1M scale: it uses a very standard GRPO formulation and doesn't yet adopt the critic models some teams have started using. He speculates large batches make gradients noisy enough-free without scaling group size, and wonders what "behaviors" means precisely — perhaps relative rankings of optimized code/performance — hoping the released environments include more detail.
More from Research
- minpts-check: one-run predictor tells if your HDBSCAN clusters survive min_samples changes — leland_mcinnes · 2026-11-12
- Architecture study sparks monitoring concerns over a multi-hop planning spike — ChowdhuryNeil · 2026-11-12
- SGLang Megakernel Pushes Muse Spark to 1,109 tok/s, 2.1x Over Baseline — hsu_byron · 2026-11-12
- ArcGrav: a cross-model Agent Control Language to cut token overhead — ArchitecturalGravity · 2026-11-12
- Open-source dual-arm robot solves tangram puzzles with 90 demos, ~60% success — DanGrover · 2026-11-12
- FutureHouse's deslopper trains Kosmos to write indistinguishably from human experts — proteinrosh · 2026-11-12