This is the pre-proceedings for the RLC 2026. You may expect minor changes.
Reinforcement Learning Journal, vol. 7, 2026, pp. TBD.
Will be presented at the Reinforcement Learning Conference (RLC), Montréal, Quebec, Canada, August 15–17, 2026.
Keywords: In-Context RL, Offline RL, Algorithm Distillation, Transformers
Existing offline in-context reinforcement learning (ICRL) methods have predominantly relied on supervised training objectives, which are known to have limitations in offline RL settings. In this study, we explore the integration of RL objectives within an offline ICRL framework. Through experiments on more than 150 GridWorld and MuJoCo environment-derived datasets, we demonstrate that optimizing RL objectives directly improves performance by approximately 30% on average compared to widely adopted Algorithm Distillation (AD), across various dataset coverages, structures, expertise levels, and environmental complexities. Furthermore, in the challenging XLand-MiniGrid environment, RL objectives doubled the performance of AD. Our results also reveal that the addition of conservatism during value learning brings additional improvements in almost all settings tested. Our findings emphasize the importance of aligning ICRL learning objectives with the RL reward-maximization goal, and demonstrate that offline RL is a promising direction for advancing ICRL.
Denis Tarasov, Alexander Nikulin, Ilya Zisman, Albina Klepach, Andrei Polubarov, Lyubaykin Nikita, Alexander Derevyagin, Igor Kiselev, and Vladislav Kurenkov. "Yes, Q-learning Helps Offline In-Context RL." Reinforcement Learning Journal, vol. 7, 2026, pp. TBD.
BibTeX:@article{tarasov2026yes,
title={Yes, Q-learning Helps Offline In-Context RL},
author={Denis Tarasov and Alexander Nikulin and Ilya Zisman and Albina Klepach and Andrei Polubarov and Lyubaykin Nikita and Alexander Derevyagin and Igor Kiselev and Vladislav Kurenkov},
journal={Reinforcement Learning Journal},
volume={7},
pages={},
year={2026}
}