Update README.md

README.md

@@ -1,6 +1,6 @@
 # AbBFN2: A flexible antibody foundation model based on Bayesian Flow Networks
 
-AbBFN2 allows for flexible task adaptation by virtue of its ability to condition the generative process on an arbitrary subset of variables. Further, since AbBFN2 is based on the Bayesian Flow Network paradigm, it can jointly model both discrete and continuous variables. Using this architecture, we provide a rich syntax which can be used to interact with the model. Regardless of conditioning information, the model generates all 45 "data modes" at inference time and arbitrary conditioning can be used to define specific tasks.
+[AbBFN2](https://www.biorxiv.org/content/10.1101/2025.04.29.651170v1) allows for flexible task adaptation by virtue of its ability to condition the generative process on an arbitrary subset of variables. Further, since AbBFN2 is based on the Bayesian Flow Network paradigm, it can jointly model both discrete and continuous variables. Using this architecture, we provide a rich syntax which can be used to interact with the model. Regardless of conditioning information, the model generates all 45 "data modes" at inference time and arbitrary conditioning can be used to define specific tasks.
 
 ## License Summary
 
@@ -221,10 +221,10 @@ If you use AbBFN2 in your research, please cite our work:
 ```bibtex
 @article{Guloglu_etal_AbBFN2,
   title={AbBFN2: A flexible antibody foundation model based on Bayesian Flow Networks},
-  author={Bora Guloglu and Miguel Bragan\c{c}a and Alex Graves and Scott Cameron and Timothy Atkinson and Liviu Copoiu},
+  author={Bora Guloglu and Miguel Bragan\c{c}a and Alex Graves and Scott Cameron and Timothy Atkinson and Liviu Copoiu and Alexandre Laterre and Thomas D Barrett},
   journal={bioRxiv},
   year={2025},
-  url={}
+  url={https://www.biorxiv.org/content/10.1101/2025.04.29.651170v1}
 }
 ```