diff --git a/adversarial_robustness/README.md b/adversarial_robustness/README.md
index e6ba2f1..5c62cb5 100644
--- a/adversarial_robustness/README.md
+++ b/adversarial_robustness/README.md
@@ -58,11 +58,32 @@ python3 eval.py \
   --ckpt=${PATH_TO_CHECKPOINT} --depth=70 --width=16 --dataset=cifar10
 ```
 
+## Generated datasets
+
+Rebuffi et al. (2021) use samples generated by a Denoising Diffusion
+Probabilistic Model [(DDPM; Ho et al., 2020)](https://arxiv.org/abs/2006.11239)
+to improve robustness. The DDPM is solely trained on the original training data
+and does not use additional external data. The following table links to datasets
+of 1M **generated** samples for CIFAR-10, CIFAR-100 and SVHN.
+
+| dataset | model | size | link |
+|---|---|:---:|:---:|
+| CIFAR-10 | DDPM | 1M | [npz](https://storage.googleapis.com/dm-adversarial-robustness/cifar10_ddpm.npz) |
+| CIFAR-100 | DDPM | 1M | [npz](https://storage.googleapis.com/dm-adversarial-robustness/cifar100_ddpm.npz) |
+| SVHN | DDPM | 1M | [npz](https://storage.googleapis.com/dm-adversarial-robustness/svhn_ddpm.npz) |
+
+To load each dataset, use NumPy. E.g.:
+
+```
+npzfile = np.load('cifar10_ddpm.npz')
+images = npzfile['image']
+labels = npzfile['label']
+```
 
 ## Citing this work
 
-If you use this code or these models in your work, please cite the relevant
-accompanying paper:
+If you use this code, data or these models in your work, please cite the
+relevant accompanying paper:
 
 ```
 @article{gowal2020uncovering,
diff --git a/adversarial_robustness/iclrw2021doing/README.md b/adversarial_robustness/iclrw2021doing/README.md
index d6cf694..e3b83aa 100644
--- a/adversarial_robustness/iclrw2021doing/README.md
+++ b/adversarial_robustness/iclrw2021doing/README.md
@@ -41,6 +41,27 @@ python3 eval.py \
   --ckpt=${PATH_TO_CHECKPOINT} --depth=70 --width=16 --dataset=cifar10
 ```
 
+## Generated datasets
+
+This work uses samples generated by a Denoising Diffusion
+Probabilistic Model [(DDPM; Ho et al., 2020)](https://arxiv.org/abs/2006.11239)
+to improve robustness. The DDPM is solely trained on the original training data
+and does not use additional external data. The following table links to datasets
+of 1M **generated** samples for CIFAR-10, CIFAR-100 and SVHN.
+
+| dataset | model | size | link |
+|---|---|:---:|:---:|
+| CIFAR-10 | DDPM | 1M | [npz](https://storage.googleapis.com/dm-adversarial-robustness/cifar10_ddpm.npz) |
+| CIFAR-100 | DDPM | 1M | [npz](https://storage.googleapis.com/dm-adversarial-robustness/cifar100_ddpm.npz) |
+| SVHN | DDPM | 1M | [npz](https://storage.googleapis.com/dm-adversarial-robustness/svhn_ddpm.npz) |
+
+To load each dataset, use NumPy. E.g.:
+
+```
+npzfile = np.load('cifar10_ddpm.npz')
+images = npzfile['image']
+labels = npzfile['label']
+```
 
 ## Citing this work
 
diff --git a/adversarial_robustness/run.sh b/adversarial_robustness/run.sh
index 0b64b41..6f11347 100755
--- a/adversarial_robustness/run.sh
+++ b/adversarial_robustness/run.sh
@@ -21,12 +21,16 @@ pip install -r adversarial_robustness/requirements.txt
 python3 -m adversarial_robustness.jax.eval \
   --ckpt=dummy \
   --dataset=cifar10 \
+  --width=1 \
+  --depth=10 \
   --batch_size=1 \
   --num_batches=1
 
 python3 -m adversarial_robustness.pytorch.eval \
   --ckpt=dummy \
   --dataset=cifar10 \
+  --width=1 \
+  --depth=10 \
   --batch_size=1 \
   --num_batches=1 \
   --nouse_cuda