pith. sign in

Arnav Raj

Identifiers

  • name variant Arnav Raj 0.60 · backfill

Papers (2)

  1. Retroactive Advantage Correction: Closed-Form V-Trace Bias Correction for Delay-Aware RLHF cs.LG · 2026 · author #1
  2. PEBS: Per-rater Empirical-Bayes Shrinkage for RLHF Reward-Model Calibration cs.LG · 2026 · author #1

Mentions

  • 2606.27580 #1 · arxiv_oai · confidence 0.70 Arnav Raj
  • 2606.27578 #1 · arxiv_oai · confidence 0.70 Arnav Raj