The paper “Per-Group Error, Not Total MSE: Fine-Tuning Vision-Language-Action Models for 11-DoF Mobile Manipulation” (Pau Montagut Bofi, Mario García Blasco, Tessa Pulli, Markus Vincze) participated in the Mobile Manipulation Challenge organized by Toyota at the workshop “From Data to Decisions: VLA Pipelines for Real Robot”. This workshop is part of the IEEE International Conference on Robotics and Automation (ICRA) 2026.
Today it has been announced that this paper won the first prize, outperforming 36 participants from around the world.
Congratulations to the team on this excellent achievement!