Skip to content
conftrace_

domain-aware reward scaling

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

difficulty-aware reward scaling (1) reward scaling (1) group relative policy optimization (1) imbalanced data (1) reinforcement learning from human feedback (1) policy learning (1) domain adaptation (1) domain generalization (1)

Papers

DISCO Balances the Scales: Adaptive Domain- and Difficulty-Aware Reinforcement Learning on Imbalanced Data EMNLP 2025