Pıer
TidesCurrentsHarbor LightsLabBottlesAshore
Pıer

Navigation

  • Tides
  • Ashore
  • Harbor Lights
  • Agent Access
  • Changelog
  • Bottles
  • Now
  • Feedback

External links

GitHubCloudborne ↗

© 2026 Pier.

WatchingResearchWatching0 independent reports0

GRPO, Dr. GRPO, and DAPO Are Three Operations on One Number: The Group-Standard-Deviation Identity

First seen · 7/2/2026, 12:00 PMLatest activity · 7/2/2026, 12:00 PM

The paper argues that GRPO, Dr. GRPO, and DAPO differ mainly in how they handle one quantity: the standard deviation of rewards within a sampled group. GRPO divides by it, Dr. GRPO removes that division, and DAPO discards groups whose deviation is zero. For binary right-or-wrong rewards, the deviation directly controls update magnitude: groups with balanced correct and incorrect answers provide the strongest signal, while unanimous groups provide none. The authors report support from the Big-Math difficulty dataset and a controlled training run.

Event heat · last 24 hours

No heat snapshots are available in the last 24 hours.

No heat snapshots are available in the last 24 hours.

Reporting Timeline

  1. AggregatorHuggingFace Daily Papers7/2, 12:00 PMnot independentRepresentative
    GRPO, Dr. GRPO, and DAPO Are Three Operations on One Number: The Group-Standard-Deviation Identity