Weighting Schedules Govern What and When Score-Based Generative Models Learn from Multimodal Data
cs.LG, cond-mat.dis-nn
Submitted: 2026-09-28
Updated: 2026-09-28
Comments: Main text : 9 pages / 5 figures Supplemental : 21 pages / 1 figure
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- The statistical thermodynamics of generative diffusion models: Phase transitions, symmetry breaking and critical instability
- Phase-aware Training Schedule Simplifies Learning in Flow-Based Generative Models
- A theory of learning data statistics in diffusion models, from easy to hard
- Scaling Rectified Flow Transformers for High-Resolution Image Synthesis
- The Generation Phases of Flow Matching: a Denoising Perspective
- Boltzmann generators for amorphous particle systems
- Efficient Diffusion Training via Min-SNR Weighting Strategy
- Elucidating the Design Space of Diffusion-Based Generative Models
- Understanding Diffusion Objectives as the ELBO with Simple Data Augmentation
- Critical windows: non-asymptotic theory for feature emergence in diffusion models
- Flow Matching for Generative Modeling
- Sora: A Review on Background, Technology, Limitations, and Opportunities of Large Vision Models
- The Interplay of Data Structure and Imbalance in the Learning Dynamics of Diffusion Models
- Autoregressive Boltzmann Generators
- Score-Based Generative Modeling through Stochastic Differential Equations
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks