逆流而上：通过反向表征对齐改进归一化流

摘要

歸一流模型（Normalizing Flows, NFs）是一類具有數學可逆架構的生成模型，其前向傳播將數據轉換至潛空間以進行密度估計，而反向傳播則從該空間生成新樣本。這一特性在表徵學習與數據生成之間建立了內在的協同效應。然而，標準歸一流模型的生成質量受限於對數似然優化所得語義表徵的不足。為解決此問題，我們提出一種新穎的對齊策略，創造性地利用歸一流模型的可逆特性：不對前向傳播進行正則化，而是將生成（反向）過程的中間特徵與強大視覺基礎模型的表徵對齊，實驗證明該方法相比簡單對齊策略具有顯著優勢。我們還引入一種無需訓練、可在測試時優化的新型分類算法，為歸一流模型內嵌的語義知識提供更本質的評估。綜合實驗表明，我們的方法使歸一流模型的訓練速度提升超過3.3倍，同時在生成質量與分類精度上均實現顯著提升。在ImageNet 64×64和256×256數據集上，本方法為歸一流模型創建了新的最優性能紀錄。代碼已開源於：https://github.com/MCG-NJU/FlowBack。

English

Normalizing Flows (NFs) are a class of generative models distinguished by a mathematically invertible architecture, where the forward pass transforms data into a latent space for density estimation, and the reverse pass generates new samples from this space. This characteristic creates an intrinsic synergy between representation learning and data generation. However, the generative quality of standard NFs is limited by poor semantic representations from log-likelihood optimization. To remedy this, we propose a novel alignment strategy that creatively leverages the invertibility of NFs: instead of regularizing the forward pass, we align the intermediate features of the generative (reverse) pass with representations from a powerful vision foundation model, demonstrating superior effectiveness over naive alignment. We also introduce a novel training-free, test-time optimization algorithm for classification, which provides a more intrinsic evaluation of the NF's embedded semantic knowledge. Comprehensive experiments demonstrate that our approach accelerates the training of NFs by over 3.3times, while simultaneously delivering significant improvements in both generative quality and classification accuracy. New state-of-the-art results for NFs are established on ImageNet 64times64 and 256times256. Our code is available at https://github.com/MCG-NJU/FlowBack.

逆流而上：通过反向表征对齐改进归一化流

Flowing Backwards: Improving Normalizing Flows via Reverse Representation Alignment

摘要

Support