Vintix II: Decision Pre-Trained Transformer is a Scalable In-Context Reinforcement Learner
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914450626314240 |
|---|---|
| author | Polubarov, Andrei Nikita, Lyubaykin Derevyagin, Alexander Grishin, Artyom Saprygin, Igor Serkov, Aleksandr Averchenko, Mark Tikhonov, Daniil Zhdanov, Maksim Nikulin, Alexander Zisman, Ilya Klepach, Albina Zemtsov, Alexey Kurenkov, Vladislav |
| author_facet | Polubarov, Andrei Nikita, Lyubaykin Derevyagin, Alexander Grishin, Artyom Saprygin, Igor Serkov, Aleksandr Averchenko, Mark Tikhonov, Daniil Zhdanov, Maksim Nikulin, Alexander Zisman, Ilya Klepach, Albina Zemtsov, Alexey Kurenkov, Vladislav |
| contents | Recent progress in in-context reinforcement learning (ICRL) has demonstrated its potential for training generalist agents that can acquire new tasks directly at inference. Algorithm Distillation (AD) pioneered this paradigm and was subsequently scaled to multi-domain settings, although its ability to generalize to unseen tasks remained limited. The Decision Pre-Trained Transformer (DPT) was introduced as an alternative, showing stronger in-context reinforcement learning abilities in simplified domains, but its scalability had not been established. In this work, we extend DPT to diverse multi-domain environments, applying Flow Matching as a natural training choice that preserves its interpretation as Bayesian posterior sampling. As a result, we obtain an agent trained across hundreds of diverse tasks that achieves clear gains in generalization to the held-out test set. This agent improves upon prior AD scaling and demonstrates stronger performance in both online and offline inference, reinforcing ICRL as a viable alternative to expert distillation for training generalist agents. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_05112 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Vintix II: Decision Pre-Trained Transformer is a Scalable In-Context Reinforcement Learner Polubarov, Andrei Nikita, Lyubaykin Derevyagin, Alexander Grishin, Artyom Saprygin, Igor Serkov, Aleksandr Averchenko, Mark Tikhonov, Daniil Zhdanov, Maksim Nikulin, Alexander Zisman, Ilya Klepach, Albina Zemtsov, Alexey Kurenkov, Vladislav Machine Learning Artificial Intelligence Robotics Recent progress in in-context reinforcement learning (ICRL) has demonstrated its potential for training generalist agents that can acquire new tasks directly at inference. Algorithm Distillation (AD) pioneered this paradigm and was subsequently scaled to multi-domain settings, although its ability to generalize to unseen tasks remained limited. The Decision Pre-Trained Transformer (DPT) was introduced as an alternative, showing stronger in-context reinforcement learning abilities in simplified domains, but its scalability had not been established. In this work, we extend DPT to diverse multi-domain environments, applying Flow Matching as a natural training choice that preserves its interpretation as Bayesian posterior sampling. As a result, we obtain an agent trained across hundreds of diverse tasks that achieves clear gains in generalization to the held-out test set. This agent improves upon prior AD scaling and demonstrates stronger performance in both online and offline inference, reinforcing ICRL as a viable alternative to expert distillation for training generalist agents. |
| title | Vintix II: Decision Pre-Trained Transformer is a Scalable In-Context Reinforcement Learner |
| topic | Machine Learning Artificial Intelligence Robotics |
| url | https://arxiv.org/abs/2604.05112 |