Research map / Image generation / Autoregressive and latent generation
DiffCollage: Parallel Generation of Large Content With Diffusion Models
Qinsheng Zhang, Jiaming Song, Xun Huang, Yongxin Chen, Ming-Yu Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use DiffCollage instead of autoregressive reconstruction | Diffusion generation on the internal landscape dataset · Internal landscape dataset | FID+: 8.97 -> 4.54 | Improved |
| Use DiffCollage instead of autoregressive replacement | Diffusion generation on the internal landscape dataset · Internal landscape dataset | Time: 14.99 -> 6.47 | Improved |
| Use DiffCollage instead of reconstruction | Image generation on LHQ 256² · LHQ 256² | FID+: 18.37 -> 16.43 | Improved |
| Use DiffCollage instead of reconstruction | Image generation on Tower 256² · Tower 256² | FID+: 19.56 -> 13.27 | Improved |
| Use DiffCollage instead of reconstruction | MDM long-duration motion generation · HumanML3D test set, 24-second motion | R Precision (top 3): 0.585 -> 0.611 | Improved |
| Use DiffCollage instead of reconstruction | MDM long-duration motion generation · HumanML3D test set, 24-second motion | FID: 1.012 -> 0.605 | Improved |
| Use DiffCollage instead of reconstruction | MDM long-duration motion generation · HumanML3D test set, 24-second motion | Multimodal Dist: 5.716 -> 5.569 | Improved |
| Use DiffCollage instead of replacement | MDM long-duration motion generation · HumanML3D test set, 24-second motion | Diversity: 9.184 -> 9.372 | Improved |
Later papers that compared with this method
2 later accepted papers report DiffCollage as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Compositional Visual Planning via Inference-Time Diffusion Scaling (ICLR 2026 · better on 7 of 8 shared results)
- DiffInfinite: Large Mask-Image Synthesis via Parallel Random Patch Diffusion in Histopathology (NeurIPS 2023 · better on 2 of 2 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- DiffSplat: Repurposing Image Diffusion Models for Scalable Gaussian Splat Generation (ICLR 2025 · similarity 0.88)
- DiffPano: Scalable and Consistent Text to Panorama Generation with Spherical Epipolar-Aware Diffusion (NeurIPS 2024 · similarity 0.88)
- DreamTime: An Improved Optimization Strategy for Diffusion-Guided 3D Generation (ICLR 2024 · similarity 0.88)
- AdaDiff: Adaptive Step Selection for Fast Diffusion Models (AAAI 2025 · similarity 0.88)
- Grid Diffusion Models for Text-to-Video Generation (CVPR 2024 · similarity 0.88)
- Compositional Discrete Latent Code for High Fidelity, Productive Diffusion Models (NeurIPS 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
