Perplexity 与 NVIDIA 发布 JOLT：师生一体的后训练强化学习方法
Perplexity and NVIDIA Researchers Introduce JOLT: a Joint Teacher-Student Reinforcement Learning Method
Perplexity 和 NVIDIA 搞了个叫 JOLT 的训练方法，一个模型当老师又当学生，在 MATH 上准度追平 GRPO，训练 token 少了 6.5 倍。
Perplexity 和 NVIDIA 提出 JOLT，一种后训练阶段的强化学习方法，让同一个模型同时充当教师和学生进行联合训练。在 MATH 基准上，JOLT 达到了与 GRPO 相当的准确率，而处理训练 token 数量只有 GRPO 的约 1/6.5。该方法针对的是传统 RL 后训练中师生模型分离带来的训练开销问题。
Perplexity and NVIDIA Researchers Introduce JOLT: a Joint Teacher-Student Reinforcement Learning Method
Perplexity and NVIDIA introduced JOLT, a post-training method that trains one model as both teacher and student. On MATH it matched GRPO’s accuracy with 6.5x fewer processed training tokens. The post Perplexity and NVIDIA Researchers Introduce JOLT: a Joint Teacher-Student Reinforcement Learning Method appeared first on MarkTechPost .
