AI Paper Atlas

Research map / Benchmarks / GEOM-Drugs

GEOM-Drugs leaderboard

10 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on GEOM-Drugs. Best validity (%): 99.7 (MolFLAE 100, NeurIPS 2025).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

GEOM-Drugs · validity (%)

10 papers, each with its best reported number for its own method; higher is better.

#validity (%) ↑MethodPaperTopic
199.7MolFLAE 100Manipulating 3D Molecules in a Fixed-Dimensional E(3)-Equivariant Latent Space
NeurIPS 2025 · Unconditional generation table
Molecules and proteins
299.49RADM-StPScaling the Prior: Size-Consistent Geometric Diffusion for 3D Molecular Generation
ICML 2026 · Table 1
Autoregressive and latent generation
399.3GeoLDM + GAGAGAGA: Gaussianity-Aware Gaussian Approximation for Efficient 3D Molecular Generation
ICLR 2026 · Table 2
Autoregressive and latent generation
499.3RADM DiT-BScalable Non-Equivariant 3D Molecule Generation via Rotational Alignment
ICML 2025 · Table 1
Autoregressive and latent generation
599.27SymDiff
DiT
SymDiff: Equivariant Diffusion via Stochastic Symmetrisation
ICLR 2025 · GEOM-Drugs results table
Autoregressive and latent generation
698.4UniGEM
EDM
UniGEM: A Unified Approach to Generation and Property Prediction for Molecules
ICLR 2025 · Generation performance table
Molecules and proteins
795.3ADiTAll-atom Diffusion Transformers: Unified generative modelling of molecules and materials
ICML 2025 · Table: Molecule generation res
Autoregressive and latent generation
892FlexiFlowFlexiFlow: decomposable flow matching for generation of flexible molecular ensemble
ICML 2026 · Table 2
Molecules and proteins
989.7GDM-aug, Energy
GDM-aug
Energy Loss Functions for Physical Systems
NeurIPS 2025 · Table: Metrics for GDM-aug on
Molecules and proteins
1064.9FMGE(3)-equivariant models cannot learn chirality: Field-based molecular generation
ICLR 2025 · Basic molecule properties tabl
Point cloud classification

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.