LATIDIA · Robótica

FlowEval: ¿Cuánta adaptación necesita un modelo de cimentación incorporado en la implementación?

arXiv:2611.11650v1 Tipo de anuncio: nuevo Resumen: El preentrenamiento a gran escala está reduciendo la cantidad de adaptación requerida para implementar modelos de cimentación incorporados, pero no se garantiza que un modelo preentrenado ya proporcione

La noticia

arXiv:2611.11650v1 Announce Type: new Abstract: Large-scale pretraining is reducing the amount of adaptation required to deploy embodied foundation models, but it is not guaranteed that a pretrained model already provide the level of performance required in every deployment. Practitioners therefore need to understand how well a pretrained model matches a target deployment and how much adaptation data is left to be collected. We study whether a small set of successful target-domain demonstrations can reveal whether and how much target-domain data will be required before any adaptation. We introduce FlowEval, a set of training-free metrics that use the flow-matching objective to probe an embodied foundation model with a set of demonstrations.