Pith. sign in

REVIEW 7 cited by

Unifying Bayesian Flow Networks and Diffusion Models through Stochastic Differential Equations

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2404.15766 v2 pith:JYL53DYU submitted 2024-04-24 cs.LG cs.AI

classification cs.LGcs.AI
keywords bfnsbayesiansamplerdifferentialdiffusionequationsfastflow
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Bayesian flow networks (BFNs) iteratively refine the parameters, instead of the samples in diffusion models (DMs), of distributions at various noise levels through Bayesian inference. Owing to its differentiable nature, BFNs are promising in modeling both continuous and discrete data, while simultaneously maintaining fast sampling capabilities. This paper aims to understand and enhance BFNs by connecting them with DMs through stochastic differential equations (SDEs). We identify the linear SDEs corresponding to the noise-addition processes in BFNs, demonstrate that BFN's regression losses are aligned with denoise score matching, and validate the sampler in BFN as a first-order solver for the respective reverse-time SDE. Based on these findings and existing recipes of fast sampling in DMs, we propose specialized solvers for BFNs that markedly surpass the original BFN sampler in terms of sample quality with a limited number of function evaluations (e.g., 10) on both image and text datasets. Notably, our best sampler achieves an increase in speed of 5~20 times for free. Our code is available at https://github.com/ML-GSAI/BFN-Solver.

Discussion (0). Sign in to comment.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Large Language Diffusion Models

    cs.CL 2025-02 unverdicted novelty 8.0 of 10

    LLaDA is a scalable diffusion-based language model that matches autoregressive LLMs like LLaMA3 8B on tasks and surpasses GPT-4o on reversal poem completion.

  2. Observation-Aligned Mask Priors for Learning Physical Dynamics from Authentic Occlusions

    cs.CV 2026-05 unverdicted novelty 7.0 of 10

    A framework pretrained on authentic binary occlusion masks uses guided sampling and intersection-based partitioning to train diffusion models on incomplete physical observations without zero-query regions.

  3. Discrete Bayesian Sample Inference for Graph Generation

    cs.LG 2025-11 unverdicted novelty 6.0 of 10

    GraphBSI uses Bayesian Sample Inference as noise-controlled SDEs to generate discrete graphs in one shot, achieving state-of-the-art results on molecular benchmarks Moses and GuacaMol.

  4. LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuning

    cs.LG 2025-05 conditional novelty 6.0 of 10

    LLaDA-V is a diffusion-based multimodal large language model that reaches competitive or state-of-the-art results on visual instruction tasks while using a non-autoregressive architecture.

  5. Efficient Long-Context Modeling in Diffusion Language Models via Block Approximate Sparse Attention

    cs.CV 2026-05 unverdicted novelty 5.0 of 10

    BA-Att introduces pre-downsampled block selection with norm-sorting and diagonal covariance correction to approximate sparse attention, yielding up to 6.95x speedup at 50% sparsity across language, multimodal, and vid...

  6. Incomplete Data, Complete Dynamics: A Diffusion Approach

    cs.LG 2025-09 unverdicted novelty 5.0 of 10

    A conditional diffusion model trained on partitioned incomplete samples for physical dynamics achieves asymptotic convergence to the true generative process under mild conditions and outperforms baselines in imputation.

  7. A Unified Measure-Theoretic View of Diffusion, Score-Based, and Flow Matching Generative Models

    cs.LG 2026-05 unverdicted novelty 4.0 of 10

    Diffusion, score-based, and flow matching models are unified as instances of learning time-dependent vector fields inducing marginal distributions governed by continuity and Fokker-Planck equations.

Pith tools