Rohan Charudatt Salvi

Identifiers

• name variant Rohan Charudatt Salvi 0.60 · backfill

Papers (1)

1. Selective-Advantage Entropy-Adaptive Horizon GRPO: Asymmetric Token-Level Discounting for Efficient Reinforcement Learning of Language Models cs.LG · 2026 · author #2

Mentions

• 2606.05434 #2 · arxiv_oai · confidence 0.70 Rohan Charudatt Salvi

Frequent Coauthors

• Chirag Chawla 1 shared papers
• Madhav S. Baidya 1 shared papers