Popular repositories Loading

1. grpo-standard-deviation-identity grpo-standard-deviation-identity Public

   Exact finite-group identity behind GRPO reward standardization, unifying GRPO / Dr. GRPO / DAPO for RLVR and LLM reasoning. Paper + code.

   TeX 4

2. no-3d-matrices no-3d-matrices Public

   Source and reproducibility code for the paper 'No 3D Matrices: A Unified Tensor-Product View of Matrix-Free Cartesian PDE Solvers' (Bay & Yearick)

   TeX 3

3. kore kore Public

   Source and reproducibility code for the paper 'Solve for the Hyperparameter, Skip the Search: Kolmogorov-Optimal Scaling Laws for Spline Regression' (Bay & Yearick)

   Python 3

4. sampling-ceilings sampling-ceilings Public

   When more sampling stops helping: a reasoning model can generate a right answer long before it can pick one. The modal and correlation ceilings of test-time scaling, with paper, figures, and code (…

   TeX 3

5. passk-single-crossing-law passk-single-crossing-law Public

   The pass@k crossover in RLVR is a theorem, not a finding: pass@k is the Laplace transform of difficulty, so sharpening forces base and RL curves to cross exactly once (Karlin variation-diminishing)…

   TeX