SOURCE-LINKED INTELLIGENCE

UniRRM: Unified Reasoning Reward Models Across Languages and Evaluation Paradigms

arXiv · AI, language, vision and robotics · article · Sep 5, 2026 · UTC

Reinforcement learning (RL) excels on tasks with verifiable rewards, but in open-ended tasks, the reliability of reward models remains a key challenge. Existing solutions either depend on costly proprietary LLM-as-a-Judge systems or opaque scalar reward models that lack interpretability. Recent works on generative reward models offer a promising alternative, but they remain constrained by static evaluation criteria, fragmented evaluation paradigms, and limited multilingual support. To address these challenges, we introduce \textbf{MixReward}, a large-scale multilingual dataset spanning six dom

Read original source ↗ Open in workspace

recordType

paper

region

Global