Force-over-time diagrams help categorize precision and force failures.
I've always been impressed by how Diffusion policies and VLAs do contact-rich tasks without force sensing.
Today I'll try to Demystify VLA Performance in Contact-Rich Tasks and How to Fix Them.
15:30 at the Act to Sense to Act Better Workshop, Hall C4, #ICRA2026
Users express excitement over the ICRA2026 workshop on VLA performance in contact-rich robotic tasks, wishing for recordings or slides due to FOMO from missing the event.
No Digg Deeper questions have been answered for this story yet.
@leto__jean would love to see a recording or your slides. FOMO of not being there!
I've always been impressed by how Diffusion policies and VLAs do contact-rich tasks without force sensing.
Today I'll try to Demystify VLA Performance in Contact-Rich Tasks and How to Fix Them.
15:30 at the Act to Sense to Act Better Workshop, Hall C4, #ICRA2026
https://active-perception-workshop.github.io
@animesh_garg We will release the paper soon with all our results.
I think we learned some overarching about contact-rich tasks and their characteristics and then mapped that to VLAs to understand their success and failure modes.
@leto__jean would love to see a recording or your slides. FOMO of not being there!