arXiv：cs.LG· Zhuoran Yang, Christopher M. Collins, Bei Peng, Luke M. Daniels, Matthew J. Rosseinsky, Vladimir V. Gusev·· 4 小时前
PXtal：在模态信息不对称下对齐粉末 X 射线衍射与晶体结构
PXtal: Learning to Align Powder X-Ray Diffraction and Crystal Structures under Information Asymmetry across Modalities
AI 导读
研究者提出 PXtal 框架，用 Unbalanced Optimal Transport（UOT）与耦合级 GKL 散度监督完整传输方案，在物理造成的信息不对称下对齐 PXRD 与晶体表示。
正文
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.
| Comments: | 22 pages, 2 figures, 14 tables |
| Subjects: | Machine Learning (cs.LG); Materials Science (cond-mat.mtrl-sci) |
| Cite as: | arXiv:2610.10653 [cs.LG] |
| (or arXiv:2610.10653v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.10653 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zhuoran Yang [view email]
[v1]
Wed, 7 Oct 2026 15:59:28 UTC (656 KB)
来源：arXiv：cs.LG · arxiv.org