AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

De Novo Molecular Generation from Mass Spectra via Many-Body Enhanced Diffusion

Xichen Sun, Wentao Wei, Jiahua Rao, Jiancong Xie, Yuedong Yang

AAAI 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Structure-based molecule design with diffusion · 3 citations (OpenAlex)

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
add many-body attention module (ablation)MBGen · NPLIB1 (molecules over 40 atoms)Tanimoto: 0.425 -> 0.525Improved
add many-body attention module (ablation)MBGen · NPLIB1 (spectra with many isomers per peak)accuracy vs isomer countImproved
add many-body attention module (ablation)MBGen · NPLIB1 (molecules under 23 atoms, under 1 isomer per peak)accuracyNo clear effect
pretrain both encoder and decoder (ablation)MBGen · NPLIB1Top-1 accuracy: 0.00% -> 12.20%Improved
pretrain decoder only vs encoder only (ablation)MBGen · NPLIB1Top-1 accuracy: 4.17% -> 8.33%Improved
add encoder pretraining on top of decoder pretraining (ablation)MBGen · NPLIB1Top-10 accuracy: 16.67% -> 22.29%Improved
replace DiffMS atom-centric diffusion with edge-centric many-body diffusionDiffMS · NPLIB1Top-1 accuracy: 8.34% -> 12.20%Improved
replace DiffMS atom-centric diffusion with edge-centric many-body diffusionDiffMS · MassSpecGymTop-1 accuracy: 2.30% -> 7.58%Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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