Hi LingBot-VA team,
Would it be possible to evaluate LingBot-VA on RMBench, a benchmark for memory-dependent robotic manipulation?
Several recent works use or cite LingBot-VA as a baseline, but the reported results differ significantly across papers. An official result would help clarify the model’s actual performance and provide a reliable reference for the community.
It would also be very helpful if you could release the corresponding training and evaluation recipe, including the checkpoint, preprocessing, training settings, and inference configuration.
Thank you for considering this request!
Hi LingBot-VA team,
Would it be possible to evaluate LingBot-VA on RMBench, a benchmark for memory-dependent robotic manipulation?
Several recent works use or cite LingBot-VA as a baseline, but the reported results differ significantly across papers. An official result would help clarify the model’s actual performance and provide a reliable reference for the community.
It would also be very helpful if you could release the corresponding training and evaluation recipe, including the checkpoint, preprocessing, training settings, and inference configuration.
Thank you for considering this request!