Research map / Probabilistic and scientific modeling / Variational and generative sampling
Riemannian Variational Flow Matching for Material and Protein Design
Olga Zaghen, Floor Eijkelboom, Alison Pouplin, Cong Liu, Max Welling, Jan-Willem van de Meent, Erik J Bekkers
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use endpoint (variational) geodesic-distance loss instead of velocity loss | RFM (intrinsic) · Spherical checkerboard (synthetic) | Coverage (%): 66.83 -> 84.21 | Improved |
| use endpoint (variational) geodesic-distance loss instead of velocity loss | RFM (intrinsic) · Hyperboloid checkerboard (synthetic) | Coverage (%): 60.75 -> 68.38 | Improved |
| use Riemannian Gaussian (geodesic) loss instead of Euclidean MSE with extrinsic prior | VFM · Spherical checkerboard (synthetic) | Coverage (%): 79.08 -> 83.10 | Improved |
| use Riemannian Gaussian loss instead of Euclidean MSE with extrinsic prior | VFM · Hyperboloid checkerboard (synthetic) | C2ST: 57.03 -> 63.55 | Hurt |
| Riemannian geometry instead of Euclidean for extrinsic models | CFM/VFM · Spherical checkerboard (synthetic) | distance to manifold: 0.044 -> 0.010 | Improved |
| variationalize the rotational loss with geodesic distance | MOFFlow · MOF structure prediction (stol=0.5, 1 sample) | match rate (%): 30.40 -> 33.52 | Improved |
| variationalize the rotational loss | MOFFlow · MOF stol=1.0, 5 samples | match rate (%): 100.0 -> 97.18 | Hurt |
| variationalize the rotational loss | QFlow · Protein backbone | designability fraction: 0.924 -> 0.968 | Improved |
| variationalize the rotational loss | ReQFlow · Protein backbone | scRMSD: 0.939 -> 0.961 (reproduced); paper 1.071 -> 0.961 | Mixed |
| set sigma_t(x)=1-t instead of 1 (ablation) | RG-VFM · synthetic checkerboards | sample quality | No clear effect |
Closest papers by meaning
- Flow Matching on General Geometries (ICLR 2024 · similarity 0.92)
- Riemannian Consistency Model (NeurIPS 2025 · similarity 0.90)
- Metric Flow Matching for Smooth Interpolations on the Data Manifold (NeurIPS 2024 · similarity 0.89)
- Categorical Flow Matching on Statistical Manifolds (NeurIPS 2024 · similarity 0.88)
- Toward a Unified Geometry Understanding : Riemannian Diffusion Framework for Graph Generation and Prediction (NeurIPS 2025 · similarity 0.87)
- Generative Modeling on Manifolds Through Mixture of Riemannian Diffusion Processes (ICML 2024 · similarity 0.87)
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.
