feat: updating model card.

Signed-off-by: Matteo Manica <[email protected]>

model_cards/article.md

@@ -10,7 +10,7 @@
 
 # Model card -- PolymerBlocks
 
-**Model Details**: *PolymerBlocks* is a sequence-based molecular generator tuned to generate blocks of polymers (e.g., catalysts and monomers). The model relies on a Variational Autoencoder architecture as described in [Born et al. (2021; *iScience*)](https://www.sciencedirect.com/science/article/pii/S2589004221002376)
+**Model Details**: *PolymerBlocks* is a sequence-based molecular generator tuned to generate blocks of polymers (e.g., catalysts and monomers). The model relies on a Variational Autoencoder architecture as described in [Born et al. (2021; *iScience*)](https://www.sciencedirect.com/science/article/pii/S2589004221002376).
 
 **Developers**: Matteo Manica and colleagues from IBM Research.
 
@@ -18,21 +18,19 @@
 
 **Model date**: Not yet published.
 
-**Model version**: Only initial model version.
+**Model version**: Only initial model version. The model has been pre-trained on 500K compounds from PubChem and further fine-tuned on the SMILES representing monomers and catalysts collected in the database presented in [Park et al. (2022)](https://doi.org/10.26434/chemrxiv-2022-811rl).
 
 **Model type**: A sequence-based molecular generator tuned to generate blocks of polymers (e.g., catalysts and monomers).
 
-**Information about training algorithms, parameters, fairness constraints or other applied approaches, and features**: 
-N.A.
+**Information about training algorithms, parameters, fairness constraints or other applied approaches, and features**: the sequence-based model is a standard GRU-based VAE trained to reconstruct SMILES representation of molecules. Given the nature of the pre-training and fine-tuning data, the model is biased to create molecules that resemble catalysts and monomers employed in ring-opening polymerization.
 
-**Paper or other resource for more information**: 
-TBD
+**Paper or other resource for more information**: Details on the model used and code can be found in [Born et al. (2021; *iScience*)](https://www.sciencedirect.com/science/article/pii/S2589004221002376).
 
 **License**: MIT
 
 **Where to send questions or comments about the model**: Open an issue on [GT4SD repository](https://github.com/GT4SD/gt4sd-core).
 
-**Intended Use. Use cases that were envisioned during development**: Chemical research, in particular drug discovery.
+**Intended Use. Use cases that were envisioned during development**: Chemical research, in particular discovery and catalysts for polymerization.
 
 **Primary intended uses/users**: Researchers and computational chemists using the model for model comparison or research exploration purposes.
 
@@ -40,7 +38,7 @@ TBD
 
 **Metrics**: N.A.
 
-**Datasets**: N.A.
+**Datasets**: See description in the model versions.
 
 **Ethical Considerations**: Unclear, please consult with original authors in case of questions.
 
@@ -49,7 +47,7 @@ TBD
 Model card prototype inspired by [Mitchell et al. (2019)](https://dl.acm.org/doi/abs/10.1145/3287560.3287596?casa_token=XD4eHiE2cRUAAAAA:NL11gMa1hGPOUKTAbtXnbVQBDBbjxwcjGECF_i-WC_3g1aBgU1Hbz_f2b4kI_m1in-w__1ztGeHnwHs)
 
 ## Citation
-TBD, temporarily please cite:
+
 ```bib
 @article{manica2022gt4sd,
   title={GT4SD: Generative Toolkit for Scientific Discovery},
@@ -57,4 +55,5 @@ TBD, temporarily please cite:
   journal={arXiv preprint arXiv:2207.03928},
   year={2022}
 }
-```
+```
+