Skip to content

Commit 1cebbb5

Browse files
committed
add new TChemGNN publication
1 parent 7ad534d commit 1cebbb5

File tree

2 files changed

+20
-0
lines changed

2 files changed

+20
-0
lines changed

_data/publications.yml

Lines changed: 20 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -14,6 +14,26 @@
1414
url = {https://arxiv.org/abs/2510.05060},
1515
}
1616
17+
- title: "Efficient Learning of Molecular Properties Using Graph Neural Networks Enhanced with Chemistry Knowledge"
18+
authors: "Tetiana Lutchyn, Marie Mardal, Benjamin Ricaud*"
19+
figure: "figs/publications/TchemGNN.png"
20+
abstract: "We build a simple GNN-based model that integrates chemistry knowledge that GNNs may have difficulties to learn. We show that this combination greatly enhances the accuracy compared with the pure GNN approach. With a simple approach, this study highlights some limitations of GNNs and the crucial benefit of giving GNN models easy access to global information about the graph in the context of applications to chemistry. We focus on regression tasks at the molecular level, on small-molecule data sets. "
21+
github: "https://github.com/uitml/TChemGNN"
22+
arxiv: "https://chemrxiv.org/engage/chemrxiv/article-details/68693d7cc1cb1ecda0442439"
23+
bibtex: |
24+
@article{doi:10.1021/acsomega.5c07178,
25+
author = {Lutchyn, Tetiana and Mardal, Marie and Ricaud, Benjamin},
26+
title = {Efficient Learning of Molecular Properties Using Graph Neural Networks Enhanced with Chemistry Knowledge},
27+
journal = {ACS Omega},
28+
volume = {10},
29+
number = {45},
30+
pages = {54421-54429},
31+
year = {2025},
32+
doi = {10.1021/acsomega.5c07178},
33+
URL = {https://doi.org/10.1021/acsomega.5c07178},
34+
eprint = {https://doi.org/10.1021/acsomega.5c07178}
35+
}
36+
1737
- title: "On Time Series Clustering with Graph Neural Networks"
1838
authors: "Jonas Berg Hansen, Andrea Cini, Filippo Maria Bianchi"
1939
venue: "TMLR 2025"

figs/publications/TchemGNN.png

224 KB
Loading

0 commit comments

Comments
 (0)