Denis Tarasov

cadrille: Multi-modal CAD Reconstruction with Reinforcement Learning

Yes, Q-learning Helps Offline In-Context RL

Vision-Language Models Unlock Task-Centric Latent Actions

N-Gram Induction Heads for In-Context RL: Improving Stability and Reducing Data Needs

Latent Action Learning Requires Supervision in the Presence of Distractors

Vintix: Action Model via In-Context Reinforcement Learning