Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.15971 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911500259557376 |
|---|---|
| author | Puniani, Cherish Kumar, Tushar Bendre, Arnav Kumar, Gaurav Singhi, Shree |
| author_facet | Puniani, Cherish Kumar, Tushar Bendre, Arnav Kumar, Gaurav Singhi, Shree |
| contents | Inspired by non-equilibrium thermodynamics, diffusion models have achieved state-of-the-art performance in generative modeling. However, their iterative sampling nature results in high inference latency. While recent distillation techniques accelerate sampling, they discard intermediate trajectory steps. This sparse supervision leads to a loss of structural information and introduces significant discretization errors. To mitigate this, we propose B-DENSE, a novel framework that leverages multi-branch trajectory alignment. We modify the student architecture to output $K$-fold expanded channels, where each subset corresponds to a specific branch representing a discrete intermediate step in the teacher's trajectory. By training these branches to simultaneously map to the entire sequence of the teacher's target timesteps, we enforce dense intermediate trajectory alignment. Consequently, the student model learns to navigate the solution space from the earliest stages of training, demonstrating superior image generation quality compared to baseline distillation frameworks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_15971 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | B-DENSE: Branching For Dense Ensemble Network Supervision Efficiency Puniani, Cherish Kumar, Tushar Bendre, Arnav Kumar, Gaurav Singhi, Shree Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Neural and Evolutionary Computing Inspired by non-equilibrium thermodynamics, diffusion models have achieved state-of-the-art performance in generative modeling. However, their iterative sampling nature results in high inference latency. While recent distillation techniques accelerate sampling, they discard intermediate trajectory steps. This sparse supervision leads to a loss of structural information and introduces significant discretization errors. To mitigate this, we propose B-DENSE, a novel framework that leverages multi-branch trajectory alignment. We modify the student architecture to output $K$-fold expanded channels, where each subset corresponds to a specific branch representing a discrete intermediate step in the teacher's trajectory. By training these branches to simultaneously map to the entire sequence of the teacher's target timesteps, we enforce dense intermediate trajectory alignment. Consequently, the student model learns to navigate the solution space from the earliest stages of training, demonstrating superior image generation quality compared to baseline distillation frameworks. |
| title | B-DENSE: Branching For Dense Ensemble Network Supervision Efficiency |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Neural and Evolutionary Computing |
| url | https://arxiv.org/abs/2602.15971 |