Sampling ideas

Diffusion step schedules, solvers, guidance and stopping rules that cut function evaluations or improve sample quality.

2 ideas found

Category: Sampling ML: Transformer Math: Stochastic processes Clear all

✓ Mechanism works 2026

Leave-One-Out Corrective Parallel Sampler

Replace standard tau-leaping in discrete diffusion generation with a first-order sampler whose per-coordinate transition is conditioned on all other current coordinates and excludes the coordinate being updated. After a parallel proposal, use the same leave-one-out conditionals to correct coordinates whose newly sampled values are inconsistent with the rest of the state, allowing large timesteps without permanently propagating simultaneous denoising errors.

Useful8/10

Difficulty5/10

Novelty7/10

Sampling Speedup Accuracy Stability

Paper: Provably adaptive sampling with uniform and remasking discrete diffusion models arXiv:2608.23554

✗ Mechanism failed 2026

Dependence-Adaptive Timestep Controller

Use an online estimate of the target distribution's remaining coordinate dependence to choose diffusion step sizes instead of using a uniform timestep grid. Take large jumps when the current state is nearly coordinate-factorized and shorten the step when the denoiser detects strong cross-coordinate dependence, targeting the paper's dependence-adaptive discretization scaling.

Useful7/10

Difficulty4/10

Novelty8/10

Sampling Speedup Sample efficiency Accuracy

Paper: Provably adaptive sampling with uniform and remasking discrete diffusion models arXiv:2608.23554