Simulation-to-real transfer learning for infrared spectroscopic chemical sensing and analysis from molecules to complex samples

2026-08-13

AI Summary

This work addresses the limitations of traditional infrared (IR) spectral analysis—reliance on manual interpretation and reference libraries—and the poor generalization, heavy annotation demands, and limited transferability of existing machine learning approaches. The authors propose UltraIR, the first large-scale foundation model for IR spectroscopy, pretrained on approximately 60 million simulated spectra. It integrates spectral reconstruction, molecular fingerprint alignment, and functional group prediction, enabling efficient chemical perception through task-specific fine-tuning. UltraIR achieves the first successful cross-domain transfer from simulated to real-world spectra and supports robust, general-purpose analysis across instruments and laboratories, even under few-shot or zero-shot settings. It significantly outperforms both conventional methods and specialized deep learning models in diverse tasks, including functional group identification, structural elucidation, mixture quantification, bacterial classification, herbal medicine authentication, microplastic detection, and soil property prediction, demonstrating exceptional generalization and data efficiency.

Abstract

Infrared (IR) spectroscopy is widely used for chemical sensing, but extracting reliable chemical information from spectra remains challenging. Conventional interpretation is labor-intensive, relies on prior knowledge and reference spectra, and is difficult to scale, whereas most machine-learning methods are tailored to individual tasks or datasets, require large labeled training sets, and transfer poorly across analytical objectives and experimental datasets. Here we introduce UltraIR, a foundation model for IR spectroscopy with more than 100 million parameters that enables simulation-to-real transfer learning for chemical sensing and analysis from molecules to complex samples. UltraIR is pretrained on approximately 60 million simulated IR spectra using spectral reconstruction, molecular fingerprint similarity alignment, and functional-group prediction, then adapted to downstream objectives with task-specific labels or targets. Across functional-group prediction, molecular structure elucidation, physicochemical property prediction, mixture-component identification and quantification, bacterial classification, medicinal-herb geographic origin traceability and constituent quantification, microplastics classification, and soil property prediction, UltraIR outperforms conventional machine-learning and task-specific deep-learning baselines. It performs strongly with limited labeled experimental spectra and in zero-shot inference for the same analytical task across Fourier-transform infrared spectrometers and laboratories, providing a route to adaptable, data-efficient chemical sensing from complex real-world samples.

Problem

• infrared spectroscopy
• chemical sensing
• transfer learning
• spectral analysis
• data efficiency

Innovation

• simulation-to-real transfer learning
• foundation model
• infrared spectroscopy
• data-efficient learning
• zero-shot inference