Search papers, labs, and topics across Lattice.
PhaseMatcher is an autoregressive framework that addresses challenges in powder X-ray diffraction phase identification by iteratively identifying phases and refining their contributions. By re-estimating phase contributions and the residual from the original observation and reference patterns after each prediction, it accounts for physically plausible variations, leading to improved complete-set identification and more accurate contribution/residual estimation compared to scalar subtraction baselines on synthetic and measured mixtures. This approach mitigates error propagation inherent in sequential phase removal, enhancing the robustness of complete phase-set recovery.
Iterative phase identification in PXRD can be significantly improved by re-estimating all phase contributions and the residual after each prediction, rather than simple subtraction.
Recovering complete phase sets from powder X-ray diffraction (PXRD) is challenging when weak-phase peaks overlap stronger signals. A natural strategy is to identify phases iteratively, removing the contribution of each identified phase from the observed pattern before predicting the next. However, even after a phase is correctly identified, misestimating its contribution can distort the residual and cause subsequent errors. We introduce PhaseMatcher, an autoregressive framework for complete phase-set identification with physics-guided spectral decomposition. After each phase prediction, PhaseMatcher re-estimates the contributions of all selected phases and the residual from the original observation and all selected reference patterns, accounting for physically plausible variation between reference patterns and the corresponding phase contributions in the observation. The resulting residual guides subsequent phase identification, while a separate stopping module determines when the phase set is complete. On synthetic mixtures and controlled mixtures constructed from measured single-phase patterns, PhaseMatcher improves complete-set identification over the evaluated baselines. On PhaseMix-135K, it also estimates contributions and residuals more accurately than scalar subtraction.
