AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Empirical Bayes Flow Matching for Continuous Cryo-EM Heterogeneity

Daniel Aibinder, Azmi Haider, Dan Rosenbaum

NeurIPS 2026 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Expressive invertible architectures · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes Empirical Bayes Flow Matching to learn generative priors over latent atomic structures from indirect, noisy cryo-EM projection images for continuous heterogeneity.

Problem

Learning generative priors over high-dimensional latent 3D atomic structures from noisy 2D cryo-EM projection images generated by a non-linear forward operator, where the structures are never directly observed.

Approach

Combines empirical Bayes with flow matching so the generative prior over latent atomic structures is learned directly from indirect, noisy projection observations rather than from clean structures; the abstract is truncated so the precise mechanism is not stated.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add persistent latents updated by stochastic approximation (ablation)EB-FM · Two-moons, measurement dim M=8manifold recoveryImproved
Add persistent latents (ablation)EB-FM · MNIST (50 measurements)EMD PSNR: 13.75 -> 15.67Improved
Use flow matching with gradient guidance instead of diffusion CG solver (DiEM-iso)EB-FM · MNIST (50 measurements)EMD PSNR: 13.75 -> 15.67Improved
Use gradient-guided EB-FM instead of DiEM-iso on two-moonsEB-FM · Two-moons under-determined regimemode preservationImproved
Remove backbone continuity loss (ablation)EB-FM · Cryo-EM 1a62_AEMD-RMSD (A): 1.62 -> 3.00Hurt
Remove backbone continuity loss (ablation)EB-FM · Cryo-EM 1egw_CEMD-RMSD (A): 6.22 -> 9.12Hurt
Randomize initialization of persistent latents (ablation)EB-FM · Cryo-EM (3 of 4 proteins)distributional metrics: 1egw_C EMD-RMSD 6.22 -> 5.56Improved
Randomize initialization of persistent latents (ablation)EB-FM · Cryo-EMpairwise RMSD / RMSF flexibility magnitude: 1a62_A pairwise RMSD 1.58 -> 1.43No clear effect
Add small pose/CTF error to all particles (ablation)EB-FM · Cryo-EM 1egw_CEMD-RMSD (A): 6.22 -> 4.73Improved
Add pose/CTF error to 30% of particles (ablation)EB-FM · Cryo-EM 1a62_AEMD-RMSD (A): 1.62 -> 2.06Hurt
Train prior jointly with latents instead of direct latent optimization alone (ablation)EB-FM · Cryo-EM latent RMSDlatent RMSD stabilityImproved
Use EB-FM prior vs cryoDRGN-guided-AF3 on EMD-RMSDEB-FM · Cryo-EM 4 proteinsEMD-RMSD: 1a62_A 1.80 -> 1.62Improved

Experimental setup

Builds on: flow matching

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