AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Teaching Molecular Dynamics to a Non-Autoregressive Ionic Transport Predictor

Jiyeon Kim, Byungju Lee, Won-Yong Shin

ICML 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Machine-learned interatomic potentials · 0 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 model-level auxiliary modality learning (trajectory knowledge transfer) (ablation)non-autoregressive predictor · Dataset 1 (Li MSD) 600KMAE log10 MSD_Li: 0.395 -> 0.344Improved
Closed-form ridge initialization instead of gradient-based optimization (ablation)predictor f1 · Dataset 1 1000KMAE log10 MSD_Li: 0.455 -> 0.402Improved
Add structure-only regularization for dual-modal trainer (ablation)dual-modal trainer g · Dataset 1 1200KMAE log10 MSD_Li: 0.408 -> 0.390Improved
Use MOMENT embedding of trajectory spectrum (ablation)trajectory embedding · Dataset 1 600KMAE log10 MSD_Li: 0.429 -> 0.344Improved
Use SevenNet structure embedding instead of LiFlow (ablation)structure embedding · Dataset 1 1000KMAE log10 MSD_Li: 0.458 -> 0.402Improved
Add polynomial expansion to embeddings (ablation)linear encoder · Dataset 1 800KMAE log10 MSD_Li: 0.409 -> 0.367Improved
Add model-level AML (pretrain on Dataset 1) (ablation)predictor f2 · Dataset 2 Na 1500KMAE log10 D_Na: 0.146 -> 0.149No clear effect
Add model-level AML (ablation)predictor f2 · Dataset 2 Na 1000KMAE log10 D_Na: 0.201 -> 0.166Improved
Add both model- and data-level AML (ablation)predictor f2 · Dataset 3 experimentalMAE log10 sigma_Li: 1.539 -> 1.388Improved
Initialize f2 encoder with W_x,T instead of W_trj (ablation)predictor f2 · Dataset 2 Na 1000KMAE log10 D_Na: 0.365 -> 0.166Improved
Non-autoregressive prediction instead of autoregressive LiFlowOurs vs LiFlow · Dataset 1inference time (s): 2910 -> 14Improved
Use trajectory-aware training rather than plain non-autoregressive modelsOurs vs Comformer · Dataset 1 1200KMAE log10 MSD_Li: 0.760 -> 0.390Improved

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