Critical Mass: Structure-Disjoint Splits Drift With the Libraries They Are Built On

arXiv:2610.12694v1 Announce Type: new Abstract: Molecular machine learning evaluates out-of-distribution performance with structure disjoint splits, a construction rule fixed once at the scale its library happened to have: threshold a pairwise distance, then assign whole connected components to folds. For predicting structures from mass spectra, the field has converged on single-linkage clustering under the myopic Maximum Common Edge Subgraph (MCES) distance at a threshold of 10. This convention was set on 29,000 molecules but public libraries now hold 227,690 molecules. MCES is NP-hard per pair over a quadratically growing set, and so the criterion has never been evaluated at that scale. We propose an optimized version of this criterion, allowing for computationally feasible application on the public libraries: Admissible bounds and a union-find skip reduce 2.6 imes 10^{10} molecular pairs to only 1.15 imes 10^{7} integer programs, a 2{,}259imes saving, while provably producing the lossless partition. We find that at this scale the graph percolates. A single component holds 95.5% of the library, capping any test set at 4.5%. What remains is not merely smaller, but systematically different: a held-out compound outweighs a training compound 88.1% of the time, because a fixed absolute budget is a tighter relative tolerance on a larger molecule. Large library benchmarks using this splitting rule naively therefore conflate structural novelty with extrapolation to larger, more complex chemistry.

Related

• Enabling Structure-Only Initialization and Out-of-Distribution Generalization in GNN-based Molecular Dynamics Simulators
• Out-of-Distribution Detection in Molecular Complexes via Diffusion Models for Irregular Graphs

Source: arXiv cs.LG | 2026-10-12