AI NEWS.ccCodeScaler-8B
We propose **CodeScaler**, an execution-free reward model designed to scale both reinforcement learning training and test-time inference for code generation. **CodeScaler** is trained on carefully curated preference data derived from verified code problems and incorporates syntax-aware code extraction and validity-preserving reward shaping to ensure stable and robust optimization. This model is the official CodeScaler-8B trained from Skywork/Skywork-Reward-V2-Qwen3-8B on LARK-Lab/CodeScalerPair-51K.
翻訳準備中のため原文を表示しています
We propose **CodeScaler**, an execution-free reward model designed to scale both reinforcement learning training and test-time inference for code generation. **CodeScaler** is trained on carefully curated preference data derived from verified code problems and incorporates syntax-aware code extraction and validity-preserving reward shaping to ensure stable and robust optimization. This model is the official CodeScaler-8B trained from Skywork/Skywork-Reward-V2-Qwen3-8B on LARK-Lab/CodeScalerPair-51K.