https://machinelearning.apple.com/research/normalizing-flows [ ] Machine Learning Research Open MenuClose Menu * Overview * Research Highlights * Publications * Events * Work with us research area Computer Vision, research area Methods and Algorithms | conference ICML content type paper | published June 2025 Normalizing Flows are Capable Generative Models AuthorsShuangfei Zhai, Ruixiang Zhang, Preetum Nakkiran, David Berthelot, Jiatao Gu, Huangjie Zheng, Tianrong Chen, Miguel Angel Bautista, Navdeep Jaitly, Josh Susskind View publication View source code (GitHub) Copy Bibtex Normalizing Flows (NFs) are likelihood-based models for continuous inputs. They have demonstrated promising results on both density estimation and generative modeling tasks, but have received relatively little attention in recent years. In this work, we demonstrate that NFs are more powerful than previously believed. We present TarFlow: a simple and scalable architecture that enables highly performant NF models. TarFlow can be thought of as a Transformer-based variant of Masked Autoregressive Flows (MAFs): it consists of a stack of autoregressive Transformer blocks on image patches, alternating the autoregression direction between layers. TarFlow is straightforward to train end-to-end, and capable of directly modeling and generating pixels. We also propose three key techniques to improve sample quality: Gaussian noise augmentation during training, a post training denoising procedure, and an effective guidance method for both class-conditional and unconditional settings. Putting these together, TarFlow sets new state-of-the-art results on likelihood estimation for images, beating the previous best methods by a large margin, and generates samples with quality and diversity comparable to diffusion models, for the first time with a stand-alone NF model. [guided_sam] Figure 1: Samples at various resolutions generated by TarFlow. [modelv3_co] Figure 2: Model architecture of TarFlow. Related readings and updates. STARFlow: Scaling Latent Normalizing Flows for High-resolution Image Synthesis June 27, 2025 | research area Computer Vision, research area Methods and Algorithms We present STARFlow, a scalable generative model based on normalizing flows that achieves strong performance in high-resolution image synthesis. The core of STARFlow is Transformer Autoregressive Flow (TARFlow), which combines the expressive power of normalizing flows with the structured modeling capabilities of Autoregressive Transformers. We first establish the theoretical universality of TARFlow for modeling continuous distributions. Building... Read more PLANNER: Generating Diversified Paragraph via Latent Language Diffusion Model November 14, 2023 | research area Speech and Natural Language Processing | conference NeurIPS Autoregressive models for text sometimes generate repetitive and low-quality output because errors accumulate during the steps of generation. This issue is often attributed to exposure bias - the difference between how a model is trained and how it is used during inference. Denoising diffusion models provide an alternative approach in which a model can revisit and revise its output. However, they can be computationally expensive, and prior... Read more Bottom banner Discover opportunities in Machine Learning. Our research in machine learning breaks new ground every day. Work with us 1. Machine Learning Research 2. Publications 3. Normalizing Flows are Capable Generative Models * Privacy Policy * Terms of Use * Legal Copyright (c) 2025 Apple Inc. All rights reserved.