[2508.08441] SpectraLLM: Uncovering the Ability of LLMs for Molecule Structure Elucidation from Multi-Spectral

Quantitative Biology > Quantitative Methods arXiv:2508.08441 (q-bio) [Submitted on 4 Aug 2025 (v1), last revised 21 Mar 2026 (this version, v2)] Title:SpectraLLM: Uncovering the Ability of LLMs for Molecule Structure Elucidation from Multi-Spectral Authors:Yunyue Su, Jiahui Chen, Zao Jiang, Zhenyi Zhong, Liang Wang, Qiang Liu, Zhaoxiang Zhang View a PDF of the paper titled SpectraLLM: Uncovering the Ability of LLMs for Molecule Structure Elucidation from Multi-Spectral, by Yunyue Su and 6 other authors View PDF Abstract:Automated molecular structure elucidation remains challenging, as existing approaches often depend on pre-compiled databases or restrict themselves to single spectroscopic modalities. Here we introduce \textbf{SpectraLLM}, a large language model that performs end-to-end structure prediction by reasoning over one or multiple spectra. Unlike conventional spectrum-to-structure pipelines, SpectraLLM represents both continuous (IR, Raman, UV-Vis, NMR) and discrete (MS) modalities in a shared language space, enabling it to capture substructural patterns that are complementary across different spectral types. We pretrain and fine-tune the model on small-molecule domains and evaluate it on four public benchmark datasets. SpectraLLM achieves state-of-the-art performance, substantially surpassing single-modality baselines. Moreover, it demonstrates strong robustness in unimodal settings and further improves prediction accuracy when jointly reasoning over diverse spec...