LATIDIA · Robótica

¿De dónde viene la capacidad física de RSI? Una ley de escala observacional de LLM como política

arXiv:2610.26848v1 Tipo de anuncio: nuevo Resumen: En este artículo, investigamos sistemáticamente los mecanismos, el comportamiento de escalado y la generalización de la auto-mejora recursiva física. Presentamos PhysRSI-eval a Quant

La noticia

arXiv:2610.26848v1 Announce Type: new Abstract: In this paper, we systematically investigate the mechanisms, scaling behavior, and generalization of physical Recursive self-improvement. We introduce PhysRSI-eval to quantify physical RSI through a five-axis metric: First-Attempt Performance, Efficiency, Success, Budget and Improvement from failure. Different VLMs are evaluated on 18 manipulation tasks under a fixed interaction budget. The analysis of experimental results motivates the Physical Barrel Hypothesis: different capabilities may constrain different parts of improvement. The five-axis metrics also distinguish systems with close public benchmark averages and show that high scores need not come with early or efficient gains. We study observational scaling from two perspectives: predicting the relative performance of VLM agents