AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Elign: Equivariant Diffusion Model Alignment from Foundational Machine Learned Force Fields

Yunyang Li, Lin Huang, Luojia Xia, Wenhe Zhang, Mark Gerstein

ICML 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Diffusion for molecular conformations

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 dense PBRS energy reward shaping (ablation)Elign · QM9validity x uniqueness (%): 93.85 -> 95.31Improved
add dense PBRS energy reward shaping (ablation)Elign · QM9molecule stability (%): 93.75 -> 93.70No clear effect
use larger MLFF UMA-1p1-M instead of UMA-1p1-S (ablation)Elign · QM9validity x uniqueness (%): 90.81 -> 95.31Improved
use larger MLFF UMA-1p1-M instead of UMA-1p1-S (ablation)Elign · QM9molecule stability (%): 94.83 -> 93.70Hurt
add force reward to energy-only reward (ablation)Elign (no PBRS) · QM9molecule stability (%): 86.00 -> 93.75Improved
add energy reward to force-only reward (ablation)Elign (no PBRS) · QM9validity x uniqueness (%): 90.21 -> 93.85Improved
add energy reward to force-only reward (ablation)Elign (no PBRS) · QM9molecule stability (%): 94.92 -> 93.75Hurt
post-train EDM with ElignEDM · QM9molecule stability (%): 82.00 -> 93.70Improved
post-train EDM with ElignEDM · QM9validity x uniqueness (%): 90.70 -> 95.31Improved
post-train EDM with Elign instead of UMA run-time guidanceEDM + UMA-1p1-M Guidance · QM9molecule stability (%): 87.25 -> 93.70Improved
post-train EDM with ElignEDM · GEOM-Drugsatom stability (%): 81.3 -> 87.94Improved
post-train EDM with ElignEDM · GEOM-DrugsRDKit validity (%): 91.9 -> 99.40Improved

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