cond-mat.mtrl-sci · 2026-05-30 · No. 12
Materials Science, 2026-05-30.
1 new papers in cond-mat.mtrl-sci. Titles, authors,
abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →
01 — The papers
1 entries-
01
What drives performance in molecular MPNNs? An operator-level factorial benchmark
Panyu Jiao, Shuizhou Chen, Yiheng Shen, Yuyang Wang, Runhai Ouyang, Wei Xie
cond-mat.mtrl-sci · cs.AI · cs.LG
Message-passing neural networks (MPNNs) are widely used for molecular property prediction, but their deployment as monolithic architectures makes it difficult to identify how specific message-passing operators affect performance. We present an operator-level factorial benchmark that decomposes 2D molecular MPNNs into the three families of message-seed initialization, node-edge fusion, and node update operators. The resulting 84 configurations...
This edition is part of The Daily Abstract — cond-mat.mtrl-sci archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.
#D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.