AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

GeomMotif: A Benchmark for Arbitrary Geometric Preservation in Protein Generation

Pavel Strashnov, Andrey Shevtsov, Viacheslav Meshchaninov, Olga Kardymon, Dmitry Vetrov

ICLR 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Structure-based 3D molecule generation · 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 structure modality to ESM3 (sequence and structure instead of sequence-only)ESM3 · GeomMotifSUN score (%): 3.5% -> 1%Hurt
scale DPLM from 650M to 3B parametersDPLM · GeomMotifSUN score (%): 2.1% -> 2.7%No clear effect
use broadly trained RFdiffusion instead of enzyme-specialised RFdiffusion2scaffolding model · GeomMotifSUN score (%): 17.9% -> 37.8%Improved
move from single-motif to paired-motif tasksRFdiffusion · GeomMotifSUN score (%): 62.4 -> 13.2Hurt
move from single-motif to paired-motif tasksESM3 (seq) · GeomMotifSUN score (%): 6.8 -> 0.1Hurt
increase fragment count from 1 to 7RFdiffusion · GeomMotif single motifsmean SUN score (%): 91.9% -> 38.7%Hurt
increase fragment count from 1 to 7Genie2 · GeomMotif single motifsmean SUN score (%): 76.7% -> 56.8%Hurt
helical content above 70% instead of below 30%Genie2 · GeomMotif single motifsSUN score (%): 42.9% -> 84.6%Improved
helical content above 70% instead of below 30%FrameFlow · GeomMotif single motifsSUN score (%): 14.0% -> 54.4%Improved
exposed low-burial motifs instead of highly buried motifs with low contact ratioGenie2 and RFdiffusion · GeomMotif single motifsSUN score (%): below 30% -> 80.7%Improved
purely helical paired motifs instead of mixed secondary structureGenie2 · GeomMotif paired motifsSUN score (%): 13.0% -> 46.5%Improved
vary TM-score novelty/uniqueness threshold between 0.7 and 0.9 (ablation)evaluation pipeline · GeomMotifmodel ranking: 0.7 to 0.9No clear effect

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