Related Experiment Video
Updated: Sep 17, 2025

On-Chip Crystallization and Large-Scale Serial Diffraction at Room Temperature
Published on: March 11, 2022
Setting new benchmarks in AI-driven infrared structure elucidation
Marvin Alberts1,2,3, Federico Zipoli1,2, Teodoro Laino1,2
1IBM Research Europe Säumerstrasse 4 8803 Rüschlikon Switzerland marvin.alberts@ibm.com.
Abstract:
Automated structure elucidation from infrared (IR) spectra represents a significant breakthrough in analytical chemistry, having recently gained momentum through the application of Transformer-based language models. In this work, we improve our original Transformer architecture, refine spectral data representations, and implement novel augmentation and decoding strategies to significantly increase performance. We report a Top-1 accuracy of 63.79% and a Top-10 accuracy of 83.95% compared to the current performance of state-of-the-art models of 53.56% and 80.36%, respectively. Our findings not only set a new performance benchmark but also strengthen confidence in the promising future of AI-driven IR spectroscopy as a practical and powerful tool for structure elucidation. To facilitate broad adoption among chemical laboratories and domain experts, we openly share our models and code.
Related Concept Videos
Infrared (IR) Spectroscopy: Overview
Different compounds display unique properties due to their...
IR Spectrometers
IR Spectrum Peak Broadening: Hydrogen Bonding
However, the extent of hydrogen bonding influences the observed stretching frequency and band broadening. Intermolecular or intramolecular...
IR Spectroscopy: Molecular Vibration Overview
Stretching vibrations are vibrational motions that occur along the bond line, changing the bond length or distance between two bonded atoms. They are further distinguished as symmetric or asymmetric. In symmetric stretching, the...
Applications of IR Spectroscopy: Overview
IR Spectrum Peak Intensity: Amount of IR-Active Bonds

