LATIDIA · Robótica

Antes de Parc Ferm\'e: Poda RL-Time para LLM incorporados eficientes en conducción autónoma

arXiv: 2605.31256v2Tipo de anuncio: reemplazar Resumen: Los modelos de lenguaje grandes incorporados (LLM) se utilizan cada vez más como módulos de razonamiento en las canalizaciones de control robótico para mejorar la interacción humano-robot, pero su memoria y g

La noticia

arXiv:2605.31256v2 Announce Type: replace Abstract: Embodied Large Language Models (LLMs) are increasingly used as reasoning modules in robotic control pipelines to improve human-robot interaction, but their memory and generation latency make real-time deployment difficult. Pruning can reduce these costs, but for controllers that undergo multiple pre- and post-training phases, the crucial question is not only how much to prune, but when pruning should occur. In this work, we propose Before Parc Ferm\'e (BPF), a pruning strategy performed during RL that compresses embodied LLM controllers while they are still being optimized for closed-loop behavior. This allows pruning decisions to account for the task-specific supervision and closed-loop feedback that shape the final