The paper investigates why reconstruction quality in latent generative models does not always predict generative performance, attributing the issue to a mismatch between encoder-induced and generation-time latent distributions. It introduces Generation‑Aware Reconstruction (GAR), a method that perturbs encoder latents with noise and denoises them through the generative model before decoding, creating a continuous trajectory that reveals how the decoder behaves across latent spaces. The resulting GAR‑FID metric correlates strongly with generation FID, and using intermediate GAR latents for decoder adaptation consistently improves generative quality across different model scales.
By Xianghong Fang, Wenjie Shu, Tongda Xu, Wenlong Mou, Dehan Kong, Tim G. J. Rudner
arXiv:2608.29335v1 Announce Type: new
Abstract: Latent generative models typically follow a two-stage pipeline, training a variational autoencoder for reconstruction and then a generative model on th...
By Guangting Zheng, Yiyuan Zhang, Tao Yang, Yunpeng Chen, Rui Zhu, Jiajun Deng, Yanyong Zhang
arXiv:2607. 18072v1 Announce Type: cross Abstract: Generative models trained on a source domain often produce samples that are poorly aligned with shifted target domains, limiting their effectiveness for target-domain data augmentation.
By Jiaqi Zhu, Xincheng Chen, Yuncheng Wu, Zhaojing Luo, Beng Chin Ooi
arXiv:2509. 05441v4 Announce Type: replace-cross Abstract: Latent generative models compress images into learned embeddings prior to synthesis, and the generation quality critically depends on how faithfully these embeddings preserve visual detail.
By Tejaswini Medi, Hsien-Yi Wang, Arianna Rampini, Margret Keuper
The paper introduces Latent Drift, a generative forecasting framework that predicts slow-evolving neurodegenerative disease progression by learning changes in a compressed semantic representation rather than full-resolution anatomy. It addresses two failure modes—identity collapse and continuous interpolation trap—by removing pixel-level identity from the prediction target and applying Finite Scalar Quantization to suppress high-frequency nuisance fluctuations. Experiments on longitudinal 3D brain MRI demonstrate that Latent Drift outperforms diffusion and autoregressive transformer baselines in both generative fidelity and clinically relevant metrics.
By Yuxiang Feng, Juncheng Wang, Chao Xu, Wenlong Hou, Huihan Wang, Yijie Qian, Yang Liu, Baigui Sun, Yong Liu, Shujun Wang
arXiv:2602. 17162v3 Announce Type: replace Abstract: Genomic Foundation Models (GFMs) typically rely on Masked Language Modeling (MLM) or Next-Token Prediction (NTP) to learn the "Laws of Nature".
By Ariel Larey, Elay Dahan, Amit Bleiweiss, Raizy Kellerman, Guy Leib, Omri Nayshool, Dan Ofer, Tal Zinger, Dan Dominissini, Gideon Rechavi, Nicole Bussola, Simon Lee, Shane O'Connell, Dung Hoang, Marissa Wirth, Alexander W. Charney, Nati Daniel, Yoli Shavit
HALO introduces a hyperspherical VAE to constrain continuous latent representations to a fixed‑radius shell, stabilizing numerical fluctuations. It then employs a masked autoregressive model that balances parallel decoding with temporal correlation learning, reducing inference steps and improving stability. Experiments show HALO achieves state‑of‑the‑art generation performance with significantly better inference efficiency compared to existing baselines.
By Chunyi Hou, Xiangfei Qiu, Hanyin Cheng, Yutong Li, Bin Yang
arXiv:2607. 01275v1 Announce Type: cross Abstract: Variational Autoencoders (VAEs) commonly assume a standard isotropic Gaussian prior over the latent space, an assumption that often fails to capture the true distribution of latent representations for complex datasets.
By Qijun Chen, Shaofan Li
arXiv:2511.21415v2 Announce Type: replace
Abstract: We introduce DiverseVAR, a framework that enhances the diversity of text-conditioned visual autoregressive models (VAR) at test time without requir...
By Mingue Park, Prin Phunyaphibarn, Phillip Y. Lee, Minhyuk Sung
MUNITE is a latent‑variable framework that unifies encoding and latent generation for any‑to‑any multimodal generation. By treating different amounts of observed evidence as varying inference problems, it models the conditional distribution over a shared latent representation, using a single conditional flow model. The approach employs self‑distillation with conditional flow matching to learn from incomplete data, achieving competitive or superior generation quality and coherence across datasets such as PolyMNIST‑D‑Q, FFHQ64, and image‑text‑audio.
By Kyeongmin Yeo, Minhyuk Sung
arXiv:2507. 23615v2 Announce Type: replace-cross Abstract: Data augmentation is becoming increasingly important across various areas of time series analysis, including forecasting, classification, and anomaly detection.
By Luis Roque, Vitor Cerqueira, Carlos Soares, Luis Torgo
arXiv:2307.00852v3 Announce Type: replace
Abstract: The natural language generation domain has witnessed great success thanks to Transformer models. Although they have achieved state-of-the-art gener...
By Yueen Ma, Dafeng Chi, Jingjing Li, Kai Song, Yuzheng Zhuang, Irwin King