Research
I'm interested in Large Behaviour Models, VLAs and improving them using RL. Most of my research is about inferring a dense reward for them and improving the ability to leverage their vast knowledge for improving exploration.
|
|
|
Coherent Soft Imitation Learning for Vision-Language-Action models
Christian Scherer,
Theo Gruner,
Daniel Palenicek,
Joe Watson,
Jan Peters
Bachelor's thesis, Nov. 2025
PDF
|
|
|
ALLMAN: A German Vision-Language-Action Model
Christian Scherer,
Maximilian Tölle,
Theo Gruner,
Daniel Palenicek,
Tim Schneider,
Patrick Schramowski,
Boris Belousov,
Jan Peters
GRC, 2025
PDF
|
|