LATIDIA · Robótica

SpaceAct: Evaluación comparativa del razonamiento espacial y temporal en 3D mediante acción de bucle cerrado y percepción activa.

arXiv:2610.38405v1 Anuncio Tipo: nuevo Resumen: Para resolver problemas en el espacio 3D en lugar de simplemente describir imágenes, los modelos de lenguaje grandes multimodales (MLLM) necesitan razonamiento espaciotemporal 3D: reconstruir geometría 3D a partir de

La noticia

arXiv:2610.38405v1 Announce Type: new Abstract: To solve problems in 3D space rather than merely describe images, multimodal large language models (MLLMs) need 3D spatiotemporal reasoning: reconstructing 3D geometry from occluded 2D views, anticipating physical dynamics, and executing precise, closed-loop actions. Prior studies test passive VQA on static images or symbolic planning, requiring no execution or supplying privileged coordinates while ignoring collision, statics, and timing. We introduce SpaceAct, an interactive benchmark evaluating 3D spatiotemporal reasoning via closed-loop physical action and active perception, with SAPA, a Spatiotemporal Active-Probing Agent framework. Built on interlocking bricks and mechanical puzzles, it has 108 verified instances across 36 task families in six categories, without privileged coordinates,