PyTorch research implementation of LiFT (Loop Flow Transformers) with adaptive recurrent depth, endpoint-preserving inference, active batching, and reproducible CPU examples.
machine-learning computer-vision deep-learning reproducible-research transformers pytorch generative-models adaptive-inference adaptive-computation flow-matching test-time-compute diffusion-transformers recurrent-transformers loop-flow-transformers flow-transformers
-
Updated
Oct 11, 2026 - Python