analiza aiqbench
This research addresses a fundamental challenge in scientific multimodal learning by developing PXtal, a framework specifically designed to handle the information asymmetry inherent in Powder X-ray Diffraction (PXRD). By leveraging Unbalanced Optimal Transport and generalized KL divergence, the model successfully aligns 1D diffraction patterns with 3D crystal structures despite significant data loss during compression. The approach demonstrates superior zero-shot transfer capabilities for materials discovery tasks, establishing a new benchmark for handling physically constrained modality mismatches in AI-driven scientific computing.
PXtal: Learning to Align Powder X-Ray Diffraction and Crystal Structures under Information Asymmetry across Modalities
arXiv:2610.10653v1 Announce Type: new Abstract: Scientific multimodal learning commonly assumes that paired views are comparably informative. Powder X-ray diffraction (PXRD) makes this mismatch explicit: compressing a three-dimensional crystal structure into a one-dimensional diffraction pattern loses information and makes the pattern harder to connect to the crystal structure that produced it. We introduce PXtal, a framework for learning aligned PXRD and crystal representations under this physically imposed information asymmetry. PXtal uses Unbalanced Optimal Transport (UOT) to adapt the cross-modal coupling and coupling-level generalized Kullback-Leibler (GKL) divergence to supervise the full transport plan. Across six test sets, including four zero-shot transfer sets, PXtal consistently outperforms the baseline models in PXRD-to-crystal candidate retrieval, with the largest gains when PXRD patterns have close but crystallographically distinct nonpaired neighbors, meaning similar input patterns associated with different crystals. The resulting crystal and PXRD encoders transfer more effectively to downstream materials and crystallographic tasks. These results identify information asymmetry as a general design problem in scientific multimodal learning: alignment objectives should reflect what each modality preserves.
Abstract:Scientific multimodal learning commonly assumes that paired views are comparably informative. Powder X-ray diffraction (PXRD) makes this mismatch explicit: compressing a three-dimensional crystal structure into a one-dimensional diffraction pattern loses information and makes the pattern harder to connect to the crystal structure that produced it. We introduce PXtal, a framework for learning aligned PXRD and crystal representations under this physically imposed information asymmetry. PXtal uses Unbalanced Optimal Transport (UOT) to adapt the cross-modal coupling and coupling-level generalized Kullback-Leibler (GKL) divergence to supervise the full transport plan. Across six test sets, including four zero-shot transfer sets, PXtal consistently outperforms the baseline models in PXRD-to-crystal candidate retrieval, with the largest gains when PXRD patterns have close but crystallographically distinct nonpaired neighbors, meaning similar input patterns associated with different crystals. The resulting crystal and PXRD encoders transfer more effectively to downstream materials and crystallographic tasks. These results identify information asymmetry as a general design problem in scientific multimodal learning: alignment objectives should reflect what each modality preserves.
Machine Learning (cs.LG); Materials Science (cond-mat.mtrl-sci)
arXiv:2610.10653 [cs.LG]
(or arXiv:2610.10653v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.10653 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 261 cuvinte
Subiecte conexe
Synthesis Superintelligence: from Semiconductors to Superconductors — Periodic Labs’ Liam Fedus and Ekin Dogus Cubuk
Latent Space · 8 oct. 2026
XRDiff: Crystal Structure Prediction from Powder X-Ray Diffraction Data Using Diffusion Models
arXiv · 15 iun. 2026
ED-CSP: Crystal Structure Prediction from Electron Diffraction
arXiv · 1 sept. 2026
PXRD-Bench: How Close Are We to Determining Crystal Structures from Powder Diffraction?
arXiv · 15 oct. 2026