双向归一化流：从数据到噪声再返回
Bidirectional Normalizing Flow: From Data to Noise and Back
BiFlow 用可学习的逆向模型替代归一化流中严格可逆的前向变换，从而摆脱因果解码瓶颈，将采样速度提升两个数量级，为基于流的生成模型提供了一种更灵活、更快的训练与采样范式。
原文依据 BiFlow learns a reverse model that approximates the underlying noise-to-data inverse mapping, enabling more flexible loss functions and architectures.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论归一化流（NF）已被确立为生成建模的原则性框架。标准 NF 由前向过程与逆向过程组成：前向过程将数据映射为噪声，逆向过程则通过反转前向过程来生成样本。典型的 NF 前向变换受显式可逆性约束，以确保逆向过程可作为其精确的解析逆。近期 TARFlow 及其变体的发展通过结合 Transformer 与自回归流复兴了 NF 方法，但也暴露出因果解码是主要瓶颈。在本工作中，我们引入双向归一化流（BiFlow），一个无需精确解析逆的框架。BiFlow 学习一个逆向模型来近似底层的噪声到数据的逆映射，从而支持更灵活的损失函数与架构。在 ImageNet 上的实验表明，与其因果解码对应方法相比，BiFlow 在提升生成质量的同时将采样速度提升高达两个数量级。BiFlow 在基于 NF 的方法中取得了最优结果，并在单次评估（“1-NFE”）方法中具有竞争力。鉴于近期 NF 领域令人鼓舞的进展，我们希望本工作能引起对这一经典范式的进一步关注。
英文原文摘要 1326 字符
Normalizing Flows (NFs) have been established as a principled framework for generative modeling. Standard NFs consist of a forward process and a reverse process: the forward process maps data to noise, while the reverse process generates samples by inverting it. Typical NF forward transformations are constrained by explicit invertibility, ensuring that the reverse process can serve as their exact analytic inverse. Recent developments in TARFlow and its variants have revitalized NF methods by combining Transformers and autoregressive flows, but have also exposed causal decoding as a major bottleneck. In this work, we introduce Bidirectional Normalizing Flow (BiFlow), a framework that removes the need for an exact analytic inverse. BiFlow learns a reverse model that approximates the underlying noise-to-data inverse mapping, enabling more flexible loss functions and architectures. Experiments on ImageNet demonstrate that BiFlow, compared to its causal decoding counterpart, improves generation quality while accelerating sampling by up to two orders of magnitude. BiFlow yields state-of-the-art results among NF-based methods and competitive performance among single-evaluation ("1-NFE") methods. Following recent encouraging progress on NFs, we hope our work will draw further attention to this classical paradigm.