HIL-UMI: Bringing Human-in-the-Loop Post-Training of Vision-Language-Action Models to Universal Manipulation Interface
A new method for human-in-the-loop post-training of vision-language-action models improves robotic manipulation capabilities.
A new method for human-in-the-loop post-training of vision-language-action models improves robotic manipulation capabilities.
Auto-scored: This is a minor capability advance in robotics with no immediate safety or misuse implications.