Upload README.md with huggingface_hub

README.md

@@ -9,11 +9,13 @@ tags:
 
 # Bad Autoencoding - Model Checkpoints
 
-Checkpoints for the paper: **"Optical Context Compression Is Just (Bad) Autoencoding"** (ICML 2025)
+Checkpoints for the paper: **"Optical Context Compression Is Just (Bad) Autoencoding"**
+
+Ivan Lee, Cheng Yang, Taylor Berg-Kirkpatrick
 
 ## Links
 
-- **Paper**: [Coming soon]
+- **Paper**: [arXiv:2512.03643](https://arxiv.org/abs/2512.03643)
 - **Code**: [https://github.com/ivnle/bad-autoencoding](https://github.com/ivnle/bad-autoencoding)
 
 ## Available Checkpoints
@@ -29,7 +31,7 @@ Checkpoints for the paper: **"Optical Context Compression Is Just (Bad) Autoenco
 - **Architecture**: DeepSeek-OCR with trainable vision encoder
 - **Image Size**: 768x768 (base)
 - **Encoder Status**: Trained (not frozen)
-- **Dataset**: 510k samples
+- **Dataset**: 510k samples from FineWiki
 
 ## Usage
 
@@ -47,10 +49,10 @@ checkpoint_path = hf_hub_download(
 ## Citation
 
 ```bibtex
-@inproceedings{bad-autoencoding2025,
+@article{lee2024optical,
   title={Optical Context Compression Is Just (Bad) Autoencoding},
-  author={...},
-  booktitle={ICML},
-  year={2025}
+  author={Lee, Ivan and Yang, Cheng and Berg-Kirkpatrick, Taylor},
+  journal={arXiv preprint arXiv:2512.03643},
+  year={2024}
 }
 ```