Popular repositories Loading
-
Sample-Smart-Not-Hard
Sample-Smart-Not-Hard PublicCode for the paper Sample Smart, Not Hard: Correctness-First Decoding for Better Reasoning in LLMs
Python 2
-
-
-
-
Confidence-as-context
Confidence-as-context PublicUse step-wise entropy or uncertainty values as persistent input to subsequent generation steps
Python
-
adaptive-head
adaptive-head PublicTrain and eval a transformer/mlp layer that generates step-wise temperature or correctness signals
Python
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.
