LATIDIA · Robótica

Evaluación del razonamiento social en modelos de visión y lenguaje incorporados

arXiv: 2610.27496v1Tipo de anuncio: nuevo Resumen: Los modelos de base ahora sirven como capas de razonamiento para los equipos de robots, pero no está claro si los modelos de lenguaje de visión especializados en la realización (VLM) preservan el área social

La noticia

arXiv:2610.27496v1 Announce Type: new Abstract: Foundation models now serve as reasoning layers for robot teams, but it remains unclear whether embodiment-specialized vision-language models (VLMs) preserve the social-reasoning capabilities of general-purpose VLMs. We evaluate Gemini Robotics-ER on SHREC, a benchmark of real human-robot social interactions, and compare it with published general-purpose VLM baselines. Robotics-ER outperforms GPT-4o-mini on error/competence detection but performs worse on detailed social-attribute judgments; adding a visual frame helps one task while hurting the other. These results suggest embodiment-specialized models may perform differently on social judgments that can be read more directly from observable interaction cues versus those requiring more inference about a person's internal state. Because the comparison