(CVPR 2026) SimLBR: Learning to Detect Fake Images by Learning to Detect Real Images 🤖🖼️🤖

This repo contains the code for the CVPR 2026 paper SimLBR. We introduce a new regularization objective, Latent Blending Regularization (LBR), for generalizable AI-generated Image Detection.

A frozen DINOv3 backbone extracts one CLS embedding per image, and a lightweight MLP head learns the real/fake classifier. When `--lbr` flag is enabled, real-image tokens are shifted assymetrically towards fake-image tokens during training, creating pseudo-fake samples near the real distribution. Validation and test always use unmodified images.

⚙️ Setup

DINOv3 loading expects DINO_V3_KEY, the API key for dinov3 model:

export DINO_V3_KEY=API_KEY_FOR_DINO_V3_L16

Run commands from the repository root:

cd ./SimLBR

🗄️ Data Layout

AIGC training uses ProGAN:

AIGCDetectionBenchMark/
  train/ProGAN/<category>/0_real/*.png
  train/ProGAN/<category>/1_fake/*.png
  test/<model>/0_real/*
  test/<model>/1_fake/*

GenImage training uses stable_diffusion_v_1_4:

GenImage/
  <model>/<category>/train/nature/*.JPEG
  <model>/<category>/train/ai/*.png
  <model>/<category>/val/nature/*.JPEG
  <model>/<category>/val/ai/*.png

Fake samples are labeled 1; real samples are labeled 0. During training, each fake anchor is paired with a random real image from the same dataset.

🏋️ Training

Baseline detector without latent blending:

python -m simlbr.train \
  --dataset_name aigc \
  --data_dir ../data/fake_data/AIGC/AIGCDetectionBenchMark \
  --train_model ProGAN \
  --val_model combined \
  --ds_fraction 0.05 \
  --batch_size 200 \
  --num_workers 20 \
  --max_epochs 5 \
  --devices 4 \
  --run_name aigc_cls_baseline

SimLBR training with CLS-token latent blending:

python -m simlbr.train \
  --dataset_name aigc \
  --data_dir ../data/fake_data/AIGC/AIGCDetectionBenchMark \
  --train_model ProGAN \
  --val_model combined \
  --ds_fraction 0.2 \
  --batch_size 200 \
  --num_workers 20 \
  --max_epochs 5 \
  --devices 4 \
  --lbr \
  --lbrdist 0.5 0.8 \
  --run_name aigc_simlbr

Fast development check:

python -m simlbr.train --fast_dev_run --wandb_mode disabled --accelerator cpu --devices 1

🕵️ Evaluation

Evaluate all subsets for a dataset:

python -m simlbr.evaluate \
  --dataset_name aigc \
  --data_dir ../data/fake_data/AIGC/AIGCDetectionBenchMark \
  --ckpt_path /path/to/checkpoint.ckpt \
  --devices 1

Evaluate selected subsets:

python -m simlbr.evaluate \
  --dataset_name aigc \
  --data_dir ../data/fake_data/AIGC/AIGCDetectionBenchMark \
  --ckpt_path /path/to/checkpoint.ckpt \
  --eval_datasets DALLE2 Midjourney

Evaluation writes evaluation_results.csv next to the checkpoint run directory. If --eval_datasets is not passed, this script launches evaluation across all generative models in the given dataset.

🎌 Important Flags

--lbr: enable Latent Blending Regularization (LBR) during training.
--lbrdist LOW HIGH: alpha range for latent blending; default is 0.5 0.8.
--hidden_layers: number of MLP hidden layers after the DINOv3 CLS token.
--activation: relu or gelu.
--dropout: dropout inside the classifier MLP.
--wandb_mode: use online, offline, or disabled.

🔖 Citation

@article{dhakal2026simlbr,
  title={SimLBR: Learning to Detect Fake Images by Learning to Detect Real Images},
  author={Dhakal, Aayush and Khanal, Subash and Sastry, Srikumar and Arndt, Jacob and Ambrozio Dias, Philipe and Lunga, Dalton and Jacobs, Nathan},
  booktitle={Computer Vision and Pattern Recognition},
  year={2026},
  organization={IEEE/CVF}
}