Softmax Transformers implement in-context RL through equivalence to weighted softmax TD updates, with error decay under contraction and parameters as global minimizers of pretraining loss.
Alphastar unplugged: Large-scale offline reinforcement learning,
5 Pith papers cite this work, alongside 5 external citations. Polarity classification is still indexing.
5
Pith papers citing it
5
external citations · external index
years
2026 5representative citing papers
With specific linear Transformer parameters, CoT generation equals iterative TD updates, yielding geometric error decay with CoT length until a context-length statistical floor, and those parameters globally minimize the pretraining loss.
A guided VAE trained on pro StarCraft replays enables four latent-space traversal strategies to produce counterfactual improvement trajectories for amateur players.
Introduces parametric open-source games as continuous analogues of program equilibria, proves equilibrium existence, and derives an exact coupling threshold for cooperation in symmetric 2x2 games under gradient ascent.
SOPE dynamically controls offline training length in online RL using actor-aligned OPE on validation data to stop when benefits saturate, achieving up to 45.6% better performance and 22x less computation on Minari tasks.
citing papers explorer
-
Beyond Linear Attention: Softmax Transformers Implement In-Context Reinforcement Learning
Softmax Transformers implement in-context RL through equivalence to weighted softmax TD updates, with error decay under contraction and parameters as global minimizers of pretraining loss.
-
Convergence and Emergence of In-Context Reinforcement Learning with Chain of Thought
With specific linear Transformer parameters, CoT generation equals iterative TD updates, yielding geometric error decay with CoT length until a context-length statistical floor, and those parameters globally minimize the pretraining loss.
-
Play Like Champions: Counterfactual Feedback Generation in Latent Space
A guided VAE trained on pro StarCraft replays enables four latent-space traversal strategies to produce counterfactual improvement trajectories for amateur players.
-
Parametric Open Source Games
Introduces parametric open-source games as continuous analogues of program equilibria, proves equilibrium existence, and derives an exact coupling threshold for cooperation in symmetric 2x2 games under gradient ascent.
-
SOPE: Stabilizing Off-Policy Evaluation for Online RL with Prior Data
SOPE dynamically controls offline training length in online RL using actor-aligned OPE on validation data to stop when benefits saturate, achieving up to 45.6% better performance and 22x less computation on Minari tasks.