Research map / Probabilistic and scientific modeling / Molecules and proteins
Periodic Bayesian Flow Networks with Additive Accuracy
Peijia Lin, Zihan Zhang, zhangrui zhao, Shaohao Rui, Junyi An, Yun-Fei Shi, Fenglei Cao, Weijie Ma, Yutong Lu
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 |
|---|---|---|---|
| add Rao-Blackwellized target (ablation) | PeriodicBFN · MP-20 | match rate: 66.11 -> 70.05 | Improved |
| add Rao-Blackwellized target (ablation) | PeriodicBFN · MP-20 | RMSE: 0.0507 -> 0.0370 | Improved |
| add 2D unit-circle embedding (ablation) | PeriodicBFN · MP-20 | match rate: 60.02 -> 70.05 | Improved |
| add 2D unit-circle embedding (ablation) | PeriodicBFN · MP-20 | RMSE: 0.0569 -> 0.0370 | Improved |
| add scalar time/accuracy input (ablation) | PeriodicBFN · MP-20 | match rate: 66.17 -> 70.05 | Improved |
| add scalar time/accuracy input (ablation) | PeriodicBFN · MP-20 | RMSE: 0.0479 -> 0.0370 | Improved |
| replace CrysBFN with PeriodicBFN | CrysBFN · MP-20 | match rate: 64.35 -> 70.05 | Improved |
| replace CrysBFN with PeriodicBFN | CrysBFN · MPTS-52 | match rate: 20.52 -> 27.03 | Improved |
| replace CrysBFN with PeriodicBFN | CrysBFN · MPTS-52 | RMSE: 0.1038 -> 0.1105 | No clear effect |
| replace DiffCSP with PeriodicBFN | DiffCSP · MP-20 | match rate: 51.49 -> 70.05 | Improved |
| add PeriodicBFN to EyeReal | EyeReal · CLF phase synthesis | PSNR: 28.86 -> 31.40 | Improved |
| add PeriodicBFN to EyeReal | EyeReal · CLF phase synthesis | time (ms): 19.9 -> 35.1 | Hurt |
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.
- FlowMM — FlowMM: Generating Materials with Riemannian Flow Matching (ICML 2024) · better on 4 of 4 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
- A Periodic Bayesian Flow for Material Generation (ICLR 2025 · similarity 0.91)
- Unified Generative Modeling of 3D Molecules with Bayesian Flow Networks (ICLR 2024 · similarity 0.86)
- Space Group Conditional Flow Matching (NeurIPS 2026 · similarity 0.84)
- Smooth Interpolation for Improved Discrete Graph Generative Models (ICML 2025 · similarity 0.84)
- Periodic Materials Generation using Text-Guided Joint Diffusion Model (ICLR 2025 · similarity 0.84)
- LLM Meets Diffusion: A Hybrid Framework for Crystal Material Generation (NeurIPS 2025 · similarity 0.83)
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.
