Sovereign 8s.is mirror of standardgalactic/correction-training
  • Python 99%
  • Shell 1%
Find a file
Repository files (latest commit first)
Filename Latest commit message Latest commit date
2026-10-14 18:59:57 -07:00
.github/workflows Publish the paper abstract with the approved teaser-first layout 2026-10-14 00:59:36 -07:00
assets Release JiT, SiT, and SiT-REPA training, sampling, and evaluation code 2026-10-14 18:57:37 -07:00
configs Release JiT, SiT, and SiT-REPA training, sampling, and evaluation code 2026-10-14 18:57:37 -07:00
docs Release JiT, SiT, and SiT-REPA training, sampling, and evaluation code 2026-10-14 18:57:37 -07:00
licenses Release JiT, SiT, and SiT-REPA training, sampling, and evaluation code 2026-10-14 18:57:37 -07:00
scripts Serve original samples as responsive SVG and JPEG 2026-10-11 16:50:47 -07:00
src Release JiT, SiT, and SiT-REPA training, sampling, and evaluation code 2026-10-14 18:57:37 -07:00
.gitignore Release JiT, SiT, and SiT-REPA training, sampling, and evaluation code 2026-10-14 18:57:37 -07:00
AGENTS.md Clarify the Markdown input used by the draft preview 2026-10-11 17:35:00 -07:00
LICENSE Release JiT, SiT, and SiT-REPA training, sampling, and evaluation code 2026-10-14 18:57:37 -07:00
README.md Restore native README method video player 2026-10-14 18:59:57 -07:00
requirements-adm.txt Release JiT, SiT, and SiT-REPA training, sampling, and evaluation code 2026-10-14 18:57:37 -07:00
requirements-eval.txt Release JiT, SiT, and SiT-REPA training, sampling, and evaluation code 2026-10-14 18:57:37 -07:00
requirements-latent.txt Release JiT, SiT, and SiT-REPA training, sampling, and evaluation code 2026-10-14 18:57:37 -07:00
requirements.txt Release JiT, SiT, and SiT-REPA training, sampling, and evaluation code 2026-10-14 18:57:37 -07:00

Correction Training for Diffusion Models

Donghoon Ahn1   Runqian Wang1   Alexei A. Efros1   Taesung Park2   Yash Belhe2
1 UC Berkeley   2 Reve

Paper: PDF arXiv: 2610.14136 Project page Checkpoints: Available

No guidance needed.

Correction achieves 2.07 FID with DiT, 1.92 with SiT, and 2.96 with JiT
on class-conditional ImageNet-256, without classifier-free guidance.

ImageNet samples generated with correction: class-conditional generation without CFG, and unconditional generation.

Correction training adds a network that refines the denoiser's prediction using the same clean target. During sampling, we reuse the corrector within each denoising step.

Method

https://github.com/user-attachments/assets/8969a92a-ced1-4862-a2b1-2dea261bac21

The corrector learns to recover the clean target from the denoiser's prediction, with stop-gradient through that prediction. At inference, its output can be corrected again before constructing the next noisy state.

Installation

This implementation supports JiT, SiT, and SiT-REPA. DiT and text-to-image code are coming soon.

Sampling requires an NVIDIA GPU. On a cluster, run the sampling commands on a GPU node.

git clone https://github.com/sunovivid/correction-training.git
cd correction-training
conda create -n correction python=3.11 pip -y --override-channels -c conda-forge # Use these flags to avoid Anaconda's ToS prompt.
conda activate correction
python -m pip install -r requirements-latent.txt

requirements-latent.txt includes requirements.txt and installs the sampling and training dependencies for all three models: JiT, SiT, and SiT-REPA. For a JiT-only setup, requirements.txt is sufficient.

Activate correction before running sampling, training, or evaluation commands.

Using venv

With Python 3.11 installed, run these commands from the repository root:

python3.11 -m venv .venv
source .venv/bin/activate
python -m pip install -r requirements-latent.txt

Sampling

Start with 16 golden retrievers in a 4 × 4 grid:

# SiT-REPA-XL/2 (try this first!)
python src/sample.py --config configs/eval/qualitative/sit-repa.yml --output samples/sit-repa-100k
# SiT-XL/2
python src/sample.py --config configs/eval/qualitative/sit.yml --output samples/sit
# JiT-L/16
python src/sample.py --config configs/eval/qualitative/jit.yml --output samples/jit

This will automatically download checkpoints from the public Release on first use. Note: Download speeds sometimes decrease significantly during downloads (< 1 MB/s). If this happens, stopping the job with Ctrl+C and then rerunning it helps.

To store checkpoints elsewhere, set export CORRECTION_CHECKPOINT_DIR=/path/to/checkpoints before running commands. The default is checkpoints/. (Allow about 11 GB of disk space per SiT/SiT-REPA checkpoint for the download parts and assembled weights.)

The paper's baseline comparisons use approximately matched NFEs. Larger K can produce better images, so we use larger K for default sampling configuration. For the paper settings, see Appendix B.1 and the evaluation guide. (Sampling configs: SiT-REPA, SiT, JiT)


You can override some options in the default configuration. Some of the main arguments you may want to modify are listed below:
Sampling arguments
Argument Role Example Default
--checkpoint Select weights or a local bundle sit-repa-50k From config
--classes ImageNet class IDs 207 250 207 (golden retriever)
--samples Total images 8 16
--batch-size Images per batch 8 4
--seed Random seed 42 0
--denoising_steps Outer denoising steps 42 From config
--corrections_per_step K in the paper: corrections per active denoising step 8 From config
--correction-interval Correction start t, end t (1 = noise, 0 = data) 0.7 0.5 From config
--cfg CFG scale; values other than 1 require K=0 3 1
--output Parent directory for timestamped runs samples/dogs samples/<checkpoint>/

See the full sampling options table for all options.


Also check out some fun examples of sweeping K and comparing correction with CFG!
Sampling examples

CLI arguments override the YAML defaults. For example:

# Siberian huskies: 16 images, a fixed seed, 4 images per batch
python src/sample.py --config configs/eval/qualitative/sit-repa.yml \
  --classes 250 --samples 16 --seed 42 --batch-size 4 --output samples/huskies

# Use the 50K checkpoint with the same sampling recipe (the default is 100K)
python src/sample.py --config configs/eval/qualitative/sit-repa.yml \
  --checkpoint sit-repa-50k --classes 207 --seed 42 --output samples/dogs-50k

# Change denoising steps, correction count K, and interval (248 NFEs)
python src/sample.py --config configs/eval/qualitative/sit-repa.yml \
  --denoising_steps 136 --corrections_per_step 4 --correction-interval 0.7 0.5 --output samples/k4

# Unconditional generation
python src/sample.py --config configs/eval/qualitative/sit-repa.yml \
  --checkpoint sit-repa-uncond-100k --unconditional --samples 16 --seed 42 \
  --output samples/unconditional

Baseline (no guidance, no correction). Use the same checkpoint with K=0 and CFG=1. These commands generate the same 16 golden retrievers with seed 0 and batch size 4, at approximately 250 NFEs:

# SiT-REPA: 250 NFEs
python src/sample.py --config configs/eval/qualitative/sit-repa.yml \
  --denoising_steps 250 --corrections_per_step 0 --cfg 1 --output samples/sit-repa-baseline

# SiT: 250 NFEs
python src/sample.py --config configs/eval/qualitative/sit.yml \
  --denoising_steps 250 --corrections_per_step 0 --cfg 1 --output samples/sit-baseline

# JiT: 249 NFEs (Heun)
python src/sample.py --config configs/eval/qualitative/jit.yml \
  --denoising_steps 125 --corrections_per_step 0 --cfg 1 --output samples/jit-baseline

Compare baseline, CFG, and correction. Run the example script:

bash src/examples/compare_sampling.sh

The script uses the same four initial noises and Siberian husky class for all three rows, at approximately 250 NFEs per image. It saves the grid to samples/compare.png.

Sweep CFG scale. Run the example script for w=1, 3, 5, 7:

bash src/examples/sweep_cfg.sh

Sweep correction count K. Run the example script for K=0, 1, 2, 4, 6 at approximately 250 NFEs per image:

bash src/examples/sweep_corrections.sh

Checkpoints

Name Model Updates Use
sit-repa-50k SiT-REPA-XL/2 50K Main-table evaluation
sit-repa-100k SiT-REPA-XL/2 100K Visual sampling
sit-repa-uncond-100k SiT-REPA-XL/2 100K Unconditional sampling
sit SiT-XL/2 100K Evaluation and sampling
jit JiT-L/16 100K Evaluation and sampling

Note that these checkpoints will be automatically downloaded when the corresponding sampling or evaluation code runs on first use (recommended). Alternatively, you can manually download it from public Release.

Bundles include both networks and their trained corrector configuration; no conversion is needed. See checkpoint settings for parameterizations, gates, and downloads.

Training and evaluation

Note

Training, checkpointing, resuming, and evaluation have been tested on short runs. Full training-to-evaluation reproduction is still in progress.

First prepare ImageNet. For an installation check, follow the two-update training and resume example on one GPU. For multi-GPU startup hangs, see distributed startup.

To train correction from the official pretrained SiT denoiser:

torchrun --standalone --nproc_per_node=8 src/train.py \
  --config configs/train/sit.yml --init-base sit \
  --data /path/to/imagenet --output runs/sit

The official base weights download automatically and are converted on load. For SiT-REPA, use --config configs/train/sit-repa.yml --init-base sit-repa. Use --init-pair sit to fine-tune our released denoiser–corrector pair.

The joint-training recipes save a restart checkpoint every 5K updates and an EMA bundle every 10K updates. Choose an output drive with about 100 GB free for a 100K run, including temporary space when replacing a restart checkpoint.

  • Data preparation: ImageNet download, preprocessing, and optional VAE caching.
  • Training: model configs, pretrained initialization, fine-tuning, and resume.
  • Evaluation: 50K generation, FID, precision, and recall.
  • Configs: model-specific training, sampling, and evaluation recipes.

Generate evaluation samples with the same command for all three models:

torchrun --standalone --nproc_per_node=8 src/evaluate.py --model sit --output runs/sit-50k

Use --model jit or --model sit-repa for the other checkpoints. The evaluation guide provides evaluator installation, reference downloads, and FID / precision / recall commands for each model.

See the documentation index for all guides.

ImageNet results

High-quality generation without CFG. Correction improves FID over the unguided baselines at approximately the same sampling budget as the CFG references.

Model Base FID ↓ Correction, no CFG ↓ CFG reference ↓
DiT-XL/2 9.58 2.07 2.51
SiT-XL/2 8.55 1.92 2.11
SiT-REPA-XL/2 5.53 1.31 1.35
JiT-L/16 14.86 2.96 2.35

ImageNet-256, FID@50K, approximately 250 network evaluations per image, including denoiser, corrector, and CFG calls.

Control refinement with K

Final SiT-REPA samples across K, with DiT FID and precision–recall curves.

Final generations share the same initial noise.

More diversity, less oversaturation than strong CFG

SiT-REPA husky samples: CFG on top, correction below, with the strongest settings highlighted.

Same class and initial noises, at approximately matched NFE. Top: CFG. Bottom: correction without CFG. Correction retains varied poses, framing, backgrounds, and colors.

Text-to-image: correction also works with CFG

Correction can reduce oversaturation and repetitive compositions while preserving variety across seeds.

CFG plus correction produces more varied bench designs and viewpoints.

Same prompt and eight seeds. Top: no correction; bottom: K=6. Left: no CFG; right: CFG (w=6).

Setting GenEval ↑ DPG-Bench ↑
CFG 68.08 83.38
CFG + correction 70.48 83.69

NanoGen-256, w=4, approximately matched sampling budget: 250 vs. 246 network evaluations.

More results include unguided text-to-image, human preferences, and additional K sweeps. Text-to-image code is coming soon.

Citation

@misc{ahn2026correction,
  title  = {Correction Training for Diffusion Models},
  author = {Ahn, Donghoon and Wang, Runqian and Efros, Alexei A. and Park, Taesung and Belhe, Yash},
  year   = {2026}
}

License

The code is released under the MIT License. Adapted components retain their original copyright and license notices in licenses/.

Acknowledgments

We build on DiT, SiT, REPA, LightningDiT, and JiT. Their original license notices are preserved in licenses/.