Spaces:
Running
Running
Commit
•
9a164e3
1
Parent(s):
c0c0555
update
Browse files
model_cards/regression_transformer_article.md
CHANGED
@@ -103,12 +103,19 @@ Model card prototype inspired by [Mitchell et al. (2019)](https://dl.acm.org/doi
|
|
103 |
## Citation
|
104 |
|
105 |
```bib
|
106 |
-
@article{
|
107 |
-
title={Regression Transformer
|
108 |
author={Born, Jannis and Manica, Matteo},
|
109 |
-
journal={
|
110 |
-
|
111 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
112 |
}
|
113 |
```
|
114 |
|
|
|
103 |
## Citation
|
104 |
|
105 |
```bib
|
106 |
+
@article{born2023regression,
|
107 |
+
title={Regression Transformer enables concurrent sequence regression and generation for molecular language modelling},
|
108 |
author={Born, Jannis and Manica, Matteo},
|
109 |
+
journal={Nature Machine Intelligence},
|
110 |
+
year={2023},
|
111 |
+
month={04},
|
112 |
+
day={06},
|
113 |
+
volume={},
|
114 |
+
number={},
|
115 |
+
pages={},
|
116 |
+
note={},
|
117 |
+
doi={10.1038/s42256-023-00639-z},
|
118 |
+
url={https://doi.org/10.1038/s42256-023-00639-z},
|
119 |
}
|
120 |
```
|
121 |
|
model_cards/regression_transformer_description.md
CHANGED
@@ -4,8 +4,8 @@
|
|
4 |
|
5 |
### Concurrent sequence regression and generation for molecular language modeling
|
6 |
|
7 |
-
The [Regression Transformer](https://
|
8 |
-
This yields a dichotomous language model that seamlessly integrates property prediction with property-driven conditional generation. For details see the [
|
9 |
|
10 |
Each `algorithm_version` refers to one trained model. Each model can be used for **two tasks**, either to *predict* one (or multiple) properties of a molecule or to *generate* a molecule (given a seed molecule and a property constraint).
|
11 |
|
|
|
4 |
|
5 |
### Concurrent sequence regression and generation for molecular language modeling
|
6 |
|
7 |
+
The [Regression Transformer](https://www.nature.com/articles/s42256-023-00639-z) is a multitask Transformer that reformulates regression as a conditional sequence modeling task.
|
8 |
+
This yields a dichotomous language model that seamlessly integrates property prediction with property-driven conditional generation. For details see the [*Nature Machine Intelligence* paper](https://www.nature.com/articles/s42256-023-00639-z), the [development code](https://github.com/IBM/regression-transformer) and the [GT4SD endpoint](https://github.com/GT4SD/gt4sd-core) for inference.
|
9 |
|
10 |
Each `algorithm_version` refers to one trained model. Each model can be used for **two tasks**, either to *predict* one (or multiple) properties of a molecule or to *generate* a molecule (given a seed molecule and a property constraint).
|
11 |
|