Update with citation for fiats.

Author: jatkinson1000

paper/paper.bib

@@ -39,6 +39,14 @@ @Online{fortitude
   year={2024},
 }
 
+@Online{fiats,
+  accessed = {2024-11-13},
+  author   = {Rouson, Damien and Rasmussen, Katherine},
+  title    = {Fiats: Functional inference and training for surrogates},
+  url      = {https://github.com/BerkeleyLab/fiats},
+  year={2024},
+}
+
 @Online{fortran-tf-lib,
   accessed = {2025-01-30},
   author   = {Cambridge-ICCS},

paper/paper.md

@@ -218,7 +218,7 @@ saving PyTorch models to TorchScript.
   However, the set of features is not as rich as PyTorch and GPU offloading
   is not currently supported.\
   The [Fiats](https://berkeleylab.github.io/fiats) (`Functional Inference And Training
-  for Surrogates') package is another approach for developing, training, and deploying
+  for Surrogates') library [@fiats] is another approach for developing, training, and deploying
   ML models directly in Fortran, with experimental GPU support at present.
 
 * **Forpy** [@forpy]\