Exploring Joint-Embedding Predictive Architectures for Self-Supervised Representation Learning of Powder X-ray Diffraction Patterns
Abstract
Powder X-ray diffraction (XRD) is the primary high-throughput method for resolving crystal structures, a cornerstone of materials science. As high-throughput synthesis and self-driving laboratories advance, automated XRD pattern analysis has become vital for accelerating materials discovery. However, traditional matrix factorization methods fail when lattice variations cause non-linear peak shifts. At the same time, supervised learning approaches face a combinatorial explosion when handling multi-phase mixtures. To overcome these challenges, we adopt a self-supervised Joint Embedding Predictive Architecture (JEPA) to learn representation embeddings from 1D diffraction patterns. By optimizing a masked self-prediction loss directly within the latent space, the framework captures high-level crystallographic semantics. Downstream evaluations show that lightweight attentive probes trained on frozen JEPA representations achieve space group classification and lattice parameter estimation accuracy close to much larger supervised vision transformers. Furthermore, latent probing reveals that the learned representation space naturally organizes physical transformations, isolating smooth trajectories for peak broadening. Interestingly, decoding along latent paths between single-phase crystal representations yields realistic multi-phase mixture patterns, demonstrating an emergent continuous structure learned without explicit multi-phase supervision.
Related
- Fibonacci and fully degenerate conical diffraction cascades: Exploring the Poincaré sphere by path interference
- Effective Graph Neural Representation of Chain Architectures of Block Copolymers Tested by Predicting RPA Instability Limits
- Representation Transitions Reveal Predictive Structure in Complex Systems: A Trajectory-Level Reconstruction in a Critical System
- Camera-in-the-loop gradient descent hologram generation for Fraunhofer diffraction
- Quantum photonic treatment of nano-polarization in diffraction gratings
BibTeX
@misc{chang2026exploring,
title = {{Exploring Joint-Embedding Predictive Architectures for Self-Supervised Representation Learning of Powder X-ray Diffraction Patterns}},
author = {Ming-Chiang Chang and Brian DeCost},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.33310},
eprint = {2610.33310},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.33310},
note = {CC0},
}
RIS
TY - UNPB
TI - Exploring Joint-Embedding Predictive Architectures for Self-Supervised Representation Learning of Powder X-ray Diffraction Patterns
AU - Ming-Chiang Chang
AU - Brian DeCost
PY - 2026
DA - 2026/10/23
PB - arXiv
DO - 10.48550/arxiv.2610.33310
UR - https://arxiv.org/abs/2610.33310
AB - Powder X-ray diffraction (XRD) is the primary high-throughput method for resolving crystal structures, a cornerstone of materials science. As high-throughput synthesis and self-driving laboratories advance, automated XRD pattern analysis has become vital for accelerating materials discovery. However, traditional matrix factorization methods fail when lattice variations cause non-linear peak shifts. At the same time, supervised learning approaches face a combinatorial explosion when handling multi-phase mixtures. To overcome these challenges, we adopt a self-supervised Joint Embedding Predictive Architecture (JEPA) to learn representation embeddings from 1D diffraction patterns. By optimizing a masked self-prediction loss directly within the latent space, the framework captures high-level crystallographic semantics. Downstream evaluations show that lightweight attentive probes trained on frozen JEPA representations achieve space group classification and lattice parameter estimation accuracy close to much larger supervised vision transformers. Furthermore, latent probing reveals that the learned representation space naturally organizes physical transformations, isolating smooth trajectories for peak broadening. Interestingly, decoding along latent paths between single-phase crystal representations yields realistic multi-phase mixture patterns, demonstrating an emergent continuous structure learned without explicit multi-phase supervision.
ER -