Data for: Training data provenance, not architecture, is the primary determinant of performance on a materials discovery benchmark
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
This dataset includes per-material predictions, variance decomposition results, error-correlation analyses, scaling analyses, and collective failure characterisations for 45 models (with 53-model sensitivity checks) on the Matbench Discovery benchmark covering 256,963 WBM structures. The analysis demonstrates that training-data provenance explains substantially more performance variance than architecture choice (partial η² of 0.84–0.89 for data versus 0.15–0.35 for architecture), a finding robust under family-aware resampling and label-permutation tests.
Fields of Study & Subjects
Machine Learning in Materials Science
Computational Drug Discovery Methods
Advanced Chemical Physics Studies
Publication details
- Journal / source
- Figshare
- Publication date
- 30 March 2026
- Language
- EN
- Document type
- Dataset