পারপ্লেক্সিটি ও NVIDIA গবেষকরা উপস্থাপন করছেন JOLT: একটি যৌথ শিক্ষক-শিক্ষার্থী রিইনফোর্সমেন্ট লার্নিং পদ্ধতি

এটি ফিডে পাওয়া অংশবিশেষ। এটি সংক্ষিপ্ত বা প্রকাশকের মূল ভাষায় হতে পারে।

Perplexity, along with NVIDIA team, has introduced JOLT, a new post-training method for language models. JOLT stands for Joint On-policy Learning and Teaching. It trains a single model in 2 roles. The teacher sees privileged information; the student does not. The key change is that the teacher is trained too, so its feedback stays usable. On MATH, JOLT matched GRPO’s accuracy with 6.5x fewer processed training tokens. For teams running RL post-training, that is a direct compute question. TL;DR What it is: a post-training method. Size: tested on Qwen3-0.6B, Qwen3-4B-Instruct, Qwen3-8B, Qwen3-30B-A3B-Instruct and DeepSeek-V2-Lite. Runs on: 1 shared model plays both roles, so no separate teacher model. Training hardware: Not disclosed. Performance: JOLT+ posts the highest mean score in every setting tested against GRPO, OPSD and π-Distill. Best: AppWorld state-test pass rate of 72.09% with JOLT+, vs 62.20% for GRPO (+9.9 points). Worst: JOLT without student reward scores 51.92% on AppWorld, below GRPO. What is Perplexity JOLT? JOLT is a self-distillation method from Perplexity and NVIDIA that trains one model as teacher and student. It builds on On-Policy Self-Distillation (OPSD). Why does RL post-training need a better teacher? Outcome-reward RL, such as GRPO, gives one reward per attempt. That signal is sparse. It does not say which tokens helped. On long agentic tasks, successes are rare, so learning is slow. On-Policy Distillation fixes sparsity by having a stronger teacher score each student token. OPSD removes the separate teacher. The same model becomes the teacher when given privileged context, such as a verified solution. The research team argues this is not enough. A teacher holding the answer can take shortcuts the student cannot follow. Its guidance may sit beyond the student’s reach. In the paper, OPSD’s teacher-student disagreement grows during MATH training. How does JOLT work? JOLT trains both roles of one shared model at the same time: