SORL trains a shortcut-model policy with a one-stage actor-critic objective and uses the learned Q-function to select among sampled actions at test time.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Scaling Offline RL via Efficient and Expressive Shortcut Models
SORL trains a shortcut-model policy with a one-stage actor-critic objective and uses the learned Q-function to select among sampled actions at test time.