MAST: Motif-Augmented Diffusion with Search Tree for Spectroscopic Molecular Structure Elucidation

Abstract

Elucidating molecular structures from spectra is a foundational problem in chemical and materials characterization, yet remains challenging due to spectral ambiguity and the vast molecular space. Although recent diffusion-based generators show strong promise for spectra-conditioned elucidation, existing methods struggle to learn robust spectra-structure relationships from limited paired data when relying solely on global spectral representation. Moreover, the repeated full sampling inference strategy incurs substantial computation overhead. To address these limitations, we propose \textbf{MAST}, a \textbf{M}otif-\textbf{A}ugmented diffusion framework with \textbf{S}earch \textbf{T}ree, for joint 2D-3D spectroscopic molecular structure elucidation. MAST introduces explicit, interpretable \emph{motif priors} as intermediate evidences throughout denoising, reducing conditional ambiguity and facilitating spectra-conditioned optimization. We further cast diffusion sampling as \emph{reward-guided tree search} to prioritize high-reward denoising trajectories, yielding a compact set of spectra-consistent candidates under limited budgets. On the QM9S multi-spectra benchmark, MAST achieves \textbf{94.89\%} exact recovery and improves 3D fidelity, while preserving high chemical validity and stability. Code is available at https://github.com/Jia040223/MAST.

Preparation, Spectroscopic Elucidation, Antioxidant Potential, Enzyme Inhibition, and Molecular Docking Studies of Novel Schiff Base-Derived Thiosemicarbazones.
Hasan Yakan, Halit Muglu, Semiha Yenigun +1
Chemistry & biodiversity30 Oct 2026Peer-reviewedCC BYFull textOrganic chemistryEurope PMCPMC
Towards Generalizable and Evidential Nuclear Magnetic Resonance-Based Molecular Structure Elucidation via Large Language Model Agent
Zheng Fang, Chen Yang, Yusen Tan +9
arXiv (Cornell University)29 Jun 2026PreprintCC BY-NC-NDFull text5 citedPhysical chemistryOpenAlex
Synthesis, Spectroscopic Characterisation, and Structural Elucidation of Ni(II) and Cu(II) Complexes of a Novel N,N’-Diaminophthalamidoxime Ligand
Emmanuel Oluwaseyi Kolawole, Oluwafunmilayo F. Adekunle
ChemRxiv9 Sep 2026PreprintCC BYFull textOrganic chemistryChemRxiv
Explainable Molecular Structure Inference from GC--MS with Diffusion Models and LLM Reranking
Changlin Liu, Tianyu Yi, Chengchun Liu +2
arXiv2 Oct 2026PreprintarXiv licenceFull textPhysical chemistry Computational physicsarXiv
Structure-Transport Relationships in Aqueous Fructose Solutions: A Combined Molecular Dynamics and NMR Diffusion Study.
San Vo Vinh Phan, Xinyue Zhou, Gabriel D Barbosa +5
The journal of physical chemistry. B1 Oct 2026Peer-reviewedCC BYFull textPhysical chemistryEurope PMC

Cite

BibTeX RIS

BibTeX

@misc{jia2026mast,
  title = {{MAST: Motif-Augmented Diffusion with Search Tree for Spectroscopic Molecular Structure Elucidation}},
  author = {Chenghao Jia and Mengdi Liu and Hong Chang and Shiguang Shan and Xilin Chen},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.12067},
  eprint = {2610.12067},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.12067},
}

RIS

TY  - UNPB
TI  - MAST: Motif-Augmented Diffusion with Search Tree for Spectroscopic Molecular Structure Elucidation
AU  - Chenghao Jia
AU  - Mengdi Liu
AU  - Hong Chang
AU  - Shiguang Shan
AU  - Xilin Chen
PY  - 2026
DA  - 2026/10/08
PB  - arXiv
DO  - 10.48550/arxiv.2610.12067
UR  - https://arxiv.org/abs/2610.12067
AB  - Elucidating molecular structures from spectra is a foundational problem in chemical and materials characterization, yet remains challenging due to spectral ambiguity and the vast molecular space. Although recent diffusion-based generators show strong promise for spectra-conditioned elucidation, existing methods struggle to learn robust spectra-structure relationships from limited paired data when relying solely on global spectral representation. Moreover, the repeated full sampling inference strategy incurs substantial computation overhead. To address these limitations, we propose \textbf{MAST}, a \textbf{M}otif-\textbf{A}ugmented diffusion framework with \textbf{S}earch \textbf{T}ree, for joint 2D-3D spectroscopic molecular structure elucidation. MAST introduces explicit, interpretable \emph{motif priors} as intermediate evidences throughout denoising, reducing conditional ambiguity and facilitating spectra-conditioned optimization. We further cast diffusion sampling as \emph{reward-guided tree search} to prioritize high-reward denoising trajectories, yielding a compact set of spectra-consistent candidates under limited budgets. On the QM9S multi-spectra benchmark, MAST achieves \textbf{94.89\%} exact recovery and improves 3D fidelity, while preserving high chemical validity and stability. Code is available at https://github.com/Jia040223/MAST.
ER  -

Abstract

Related

Cite