JOLT reduce los costos de entrenamiento en 6,5 veces mientras mantiene la precisión, lo que hace que el aprendizaje por refuerzo de modelos de lenguaje a gran escala sea significativamente más eficiente y asequible.
Resumen generado con IA. Consulta el original para comprobar el contexto y la precisión.
Este es el extracto disponible del feed. Puede estar abreviado o en el idioma original.
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:
Una ayuda automática para ordenar, basada en fuentes, palabras clave y entidades, en una escala de 1–12. Alta: 9–12; media: 6–8; baja: 1–5. No mide urgencia, verifica noticias ni representa respaldo editorial.
La traducción se está preparando.
Granite Speech 5.0 2B, de código abierto de IBM, permite una transcripción y traducción multilingüe precisas, ofreciendo a los desarrolladores capacidades flexibles de procesamiento de voz en el dispositivo.
Esto es importante porque valida si un modelo de voz diminuto y compatible con CPU puede superar genuinamente a competidores más grandes, permitiendo una transcripción eficiente y de bajo costo para los desarrolladores.