Sharpen Before You Adapt: Data-Free Entry-State Sharpening for Test-Time Reinforcement Learning
Test-time reinforcement learning (TTRL) adapts language models on unlabeled test problems using supervision derived from their own samples. This makes the checkpoint's \emph{entry state} consequential: a diffuse policy provides noisier self-supervision and may spend much of a limited adaptation budget merely concentrat...
Zhan-Ming Zhang, Vinoth Selvendran
· 0 citations