Research map / Image generation / Fast diffusion sampling
Condition Errors Refinement in Autoregressive Image Generation with Diffusion Loss
Yucheng Zhou, Hao Li, Jianbing Shen
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 |
|---|---|---|---|
| add OT-based condition refinement (Wasserstein gradient flow) to autoregressive diffusion-loss model (ablation) | Baseline (AR) · ImageNet 256x256 | FID: 2.02 -> 1.52 | Improved |
| add OT-based condition refinement to MAR | MAR 943M · ImageNet 256x256 | FID: 1.55 -> 1.31 | Improved |
| add OT-based condition refinement to MAR (ablation) | MAR 208M · ImageNet 256x256 | FID: 2.31 -> 1.96 | Improved |
| add OT-based condition refinement to MAR (ablation) | MAR 479M · ImageNet 256x256 | FID: 1.78 -> 1.59 | Improved |
| add OT-based condition refinement to MAR at higher resolution | MAR ~481M · ImageNet 512x512 | FID: 1.73 -> 1.58 | Improved |
| use autoregressive condition generation instead of static conditional diffusion (ablation) | conditional diffusion baseline · ImageNet 256x256 | FID: 3.26 -> 2.02 | Improved |
| compare Ours (AR) with De-MAR and RAR | Ours (AR) · ImageNet 256x256 | FID: 1.52 vs De-MAR 1.47, RAR 1.50 | Mixed |
| add OT condition refinement | denoising process · ImageNet 256x256 | SNR during denoising | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DiffiT — DiffiT: Diffusion Vision Transformers for Image Generation (ECCV 2024) · better on 1 of 1 shared result
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
- Denoising Autoregressive Transformers for Scalable Text-to-Image Generation (ICLR 2025 · similarity 0.89)
- Autoregressive Distillation of Diffusion Transformers (CVPR 2025 · similarity 0.89)
- NoiseCtrl: A Sampling-Algorithm-Agnostic Conditional Generation Method for Diffusion Models (CVPR 2025 · similarity 0.88)
- EVODiff: Entropy-aware Variance Optimized Diffusion Inference (NeurIPS 2025 · similarity 0.88)
- Autoregressive Image Generation without Vector Quantization (NeurIPS 2024 · similarity 0.88)
- Revisiting Non-Autoregressive Transformers for Efficient Image Synthesis (CVPR 2024 · 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.
