TY  - RPRT
TI  - Towards Scalable Multi-Task Reinforcement Learning with Large Decision Models
AU  - Thibaut Kulak
PY  - 2026
UR  - https://arxiv.org/abs/2606.24962
ID  - 2606.24962
ER  -