vcrl-variance-curriculum-rl

【效率与协作】VCRL improves LLM mathematical reasoning by dynamically adjusting training sample difficulty based on group reward variance. Uses variance-based sampling with memory replay to focus on moderately-difficult samples where models succeed ~50% of the time, achieving 4.67-point improvement over GRPO on 8B models.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准