AI Paper Atlas

Research map / Large language models / Agents, code and math

SpectraLLM: Uncovering the Ability of LLMs for Molecule Structure Elucidation from Multi-Spectra

Yunyue Su, Jiahui Chen, Zao Jiang, Zhenyi Zhong, Liang Wang, Qiang Liu, Zhaoxiang Zhang

ICLR 2026 · Large language models › Agents, code and math · approach: LLMs for scientific discovery

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
fuse IR, Raman and UV-Vis spectra in one prompt instead of single spectrum (ablation)SpectraLLM · QM9sTanimoto (ECFP4): 0.2500 -> 0.3355 (Raman alone vs joint)Improved
fuse IR, Raman and UV-Vis spectra (ablation)SpectraLLM · QM9sMCES: more than 10 -> 4.9647Improved
fuse 1H, 13C and HSQC NMR instead of one NMR type (ablation)SpectraLLM · Multimodal SpectroscopicTanimoto (ECFP4): 0.2058 (HSQC alone) -> 0.4151Improved
add IR and MS to NMR input (ablation)SpectraLLM · Multimodal SpectroscopicTanimoto (ECFP4): 0.4151 -> 0.4875Improved
add IR to NMR input (without MS) (ablation)SpectraLLM · Multimodal SpectroscopicTanimoto (ECFP4): 0.4151 -> 0.4121No clear effect
add MS to IR input (ablation)SpectraLLM · Multimodal SpectroscopicTanimoto (ECFP4): 0.1844 (MS) / 0.1720 (IR) -> 0.2300Improved
use LLM text-based spectra representation vs specialized baselines on NMRSpectraLLM vs NMR2Struct · Multimodal Spectroscopic NMRTanimoto (ECFP4): 0.0433 -> 0.4151Improved
use SpectraLLM vs baselines on UV-Vis aloneSpectraLLM · QM9s UV-VisTanimoto (ECFP4): 0.0965/0.0716 -> 0.0790 (modest)Improved
use SpectraLLM vs DiffMS on MassSpecGymSpectraLLM · MassSpecGymTanimoto (ECFP4): 0.1597 (DiffMS) vs 0.1533No clear effect
use SpectraLLM vs baselines on MassBankSpectraLLM · MassBankfunctional group overlap: 0.2999 -> 0.4539Improved
train on multi-spectral inputs then test on unseen single-spectrum (ablation)SpectraLLM · QM9s (cross-domain)generalizationImproved
replace text-based LLM with vision-language models (ablation)SpectraLLM · structure elucidationperformanceHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy