Apple's NTM: exact-likelihood few-step diffusion via conditional normalizing flows
Apple ML Research · rss · 2026-10-08
Apple ML Research published Normalizing Trajectory Models (NTM). Diffusion models assume sampling decomposes into many small Gaussian denoising steps — an assumption that breaks when generation is compressed to a few coarse transitions. Existing few-step methods rely on distillation, consistency training, or adversarial objectives, sacrificing the likelihood framework. NTM instead models each reverse step as an expressive conditional normalizing flow with exact likelihood training, combining shallow invertible blocks within each step and deep parallel structure across steps.
More from Research
- Hugging Face launches Robotic Episodes Viewer for 24k+ LeRobot datasets — mishig25 · 2026-10-09
- Blind humanoid walks, plays soccer and lifts suitcases with joint encoders only — accepted at Humanoids 2026 — Jan_R_Peters · 2026-10-09
- Delete object info from observations and PPO learns to search anyway — TU Darmstadt on its Humanoids 2026 paper — Jan_R_Peters · 2026-10-09
- U-Space finds an interpretable subspace for LLM uncertainty, no training needed — Tobias Braun · 2026-10-09
- CARE certifies VLA inference speedups up to 10.8x with statistical guarantees — UMCP · 2026-10-09
- SOL: a sample-based distributional metric proposed for evaluating text diffusion LMs — NandoDF · 2026-10-09