Do Heavy Tails Help Diffusion? On the Subtle Trade-off Between Initialization and Training
cs.LG
Submitted: 2026-05-13
Updated: 2026-11-12
Comments: 29 pages, 13 figures/tables. Accepted at NeurIPS 2026
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Generalizing Score-based generative models for Heavy-tailed Distributions
- Diffusion Generative Models Meet Compressed Sensing, with Applications to Imaging and Finance
- Lipschitz regularity in Flow Matching and Diffusion Models: sharp sampling rates and functional inequalities
- Generalization bounds for score-based generative models: a synthetic proof
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis