GenFirst：生成先于重建的稳定端到端潜在生成建模
原标题：Paper page - GenFirst: Generation Before Reconstruction for Stable End-to-End Latent Generative Modeling
AI 摘要
GenFirst 提出了一种生成先于重建的端到端潜在生成模型训练策略，通过熵保持和非对称动力学避免潜在空间坍缩，实现了稳定的直接端到端训练。该方法在 ImageNet-256 上 SiT 模型达到 gFID 0.97（带 CFG）和 1.45（不带 CFG），MMDiT 在文本到图像生成中 GenEval 得分 0.90，并扩展到统一多模态生成和表示学习。
正文节选
GenFirst: Generation Before Reconstruction for Stable End-to-End Latent Generative Modeling Abstract Direct end-to-end training of latent generative models avoids collapse via entropy preservation and asymmetric dynamics, using a generation-first strategy to achieve state-of-the-art image synthesis and unified multimodal generation. Latent generative models typically follow a two-stage pipeline, training a variational autoencoder for reconstruction and then a generative model on the frozen laten