Generative modeling and self-supervised representation learning (SSL) optimize structurally different objectives: generative training rewards distributional fidelity, while SSL rewards semantic coherence. Yet recent work repeatedly finds that SSL features improve generative training, though the mechanism of this synergy remains unclear. Here, we study the benefits of SSL in generative modeling in the framework of one-step generation where the role of representation is explicit: frozen SSL features are used to match generated samples to real data. We use the Sinkhorn divergence in that feature space, providing a tractable surrogate for the Wasserstein distance, the population-level discrepancy approximated by Fréchet-style evaluation metrics (such as FID). We find that this objective becomes highly effective when computed in a semantically structured SSL feature space (a 39$\times$ reduction in ImageNet FID). We trace this behavior primarily to matching estimation: semantic SSL features that suppress nuisance reconstruction details induce a more compact geometry, making distribution matching more tractable. As a consequence, the best training SSL features need not match the features used by the evaluation metric. In particular, we show that using Inception as the feature extractor can improve FID while degrading matching stability and sample quality, revealing a form of metric hacking. Using extensive experiments on ImageNet, we identify which SSL feature families lead to best generation performance and show that matching stability is a quantitative criterion for selecting them. Code is available at https://github.com/Genentech/semantic-transport-generation.
Scan2LoD3: Reconstructing semantic 3D building models at LoD3 using ray casting and Bayesian networks
Technical University of Munich(慕尼黑工业大学);University of Oxford(牛津大学);Bruno Kessler Foundation(布鲁诺·凯斯勒基金会);Munich University of Applied Sciences(慕尼黑应用技术大学)
RangeUDF: Semantic Surface Reconstruction from 3D Point Clouds
University of Oxford(牛津大学);The Hong Kong Polytechnic University(香港理工大学);Durham University(杜伦大学);Nanjing University of Aeronautics and Astronautics(南京航空航天大学);IIAI
A Cycle Ride to HDR: Semantics Aware Self-Supervised Framework for Unpaired LDR-to-HDR Image Reconstruction
School of Information Technology, Monash University Malaysia(莫纳什大学马来西亚分校信息科技学院);Faculty of Information Technology, Monash University(莫纳什大学信息科技学院);Robotics and Autonomous Systems Lab, TCS Research(TCS研究机器人与自主系统实验室)
A Channel-Triggered Backdoor Attack on Wireless Semantic Image Reconstruction
School of Telecommunications Engineering, Xidian University(西安电子科技大学通信工程学院);Department of Computing, Hong Kong Polytechnic University(香港理工大学计算学系);Department of Electrical and Computer Engineering, University of Waterloo(滑铁卢大学电气与计算机工程系)