Upload README.md with huggingface_hub

README.md

@@ -14,6 +14,17 @@ A 2.6 B parameter image-to-video diffusion model conditioned on a per-frame
 camera trajectory, paired with the LTX-2 sink-bidirectional Euler refiner
 for high-fidelity decoding.
 
+Paper: <https://arxiv.org/abs/2605.15178>
+
+```bibtex
+@article{sanawm,
+  title  = {SANA-WM: Camera-Controlled Image-to-Video Diffusion},
+  author = {NVIDIA},
+  journal = {arXiv preprint arXiv:2605.15178},
+  year   = {2026}
+}
+```
+
 | Component                  | Path in repo                              | Size  |
 |----------------------------|-------------------------------------------|-------|
 | Sana DiT (Stage 1)         | `dit/sana_wm_1600m_720p.safetensors`      | 10 GB |