← Search

Ehsan Kamalinejad

2 accepted papers

2026

G$^2$RPO: Geometric GRPO; Escaping LLM's Reasoning Rut to Break Accuracy--Entropy Trade-off

ICML 2026poster

Reinforcement learning with verifiable rewards (RLVR) is a cornerstone of post-training for large reasoning models, yet widely used algorithms such as Group Relative Policy Optimization (GRPO) often exhibit \textbf{diversity collapse}. We provide a geometric diagnosis by formalizing GRPO as a dynami…

Cited by 0SourceScholar
2026

Rate or Fate? RLV$^{\varepsilon}$R: Reinforcement Learning with Verifiable Noisy Rewards

ICML 2026spotlight

Reinforcement learning with verifiable rewards (RLVR) trains a policy by verifying sampled completions and reinforcing higher-scoring outputs, but practical verifiers (e.g., incomplete unit tests or noisy judges) are prone to false positives and false negatives. We ask when such noise merely slows l…

Cited by 0SourceScholar