面向人体视频生成的高效训练：熵引导的优先渐进学习
Efficient Training for Human Video Generation with Entropy-Guided Prioritized Progressive Learning
提出条件熵膨胀（CEI）来量化模型各组件对条件生成任务的重要性，并据此安排训练优先级；同时用收敛效率自适应地提升训练复杂度。这套「按重要性排序 + 渐进加难」的训练调度可直接迁移到高分辨率视频扩散模型或大分辨率图像扩散模型的训练中，以降低显存与时间开销。
原文依据 achieving up to 2.2x training speedup and 2.4x GPU memory reduction without compromising generative performance
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论随着扩散模型的发展，人体视频生成取得了快速进展，但在高分辨率、多帧数据上训练这类模型所伴随的高计算成本与大量显存消耗构成了显著挑战。本文提出熵引导的优先渐进学习（Ent-Prog），一个面向人体视频生成扩散模型的高效训练框架。首先，我们引入条件熵膨胀（CEI）来评估不同模型组件对目标条件生成任务的重要性，从而对最关键组件进行优先训练。其次，我们引入自适应渐进调度，通过度量收敛效率在训练过程中自适应地提升计算复杂度。Ent-Prog 在保持模型性能的同时降低了训练时间与 GPU 显存消耗。在三个数据集上的大量实验证明了 Ent-Prog 的有效性，在不损害生成性能的前提下实现了最高 2.2 倍的训练加速与 2.4 倍的 GPU 显存降低。
英文原文摘要 1111 字符
Human video generation has advanced rapidly with the development of diffusion models, but the high computational cost and substantial memory consumption associated with training these models on high-resolution, multi-frame data pose significant challenges. In this paper, we propose Entropy-Guided Prioritized Progressive Learning (Ent-Prog), an efficient training framework tailored for diffusion models on human video generation. First, we introduce Conditional Entropy Inflation (CEI) to assess the importance of different model components on the target conditional generation task, enabling prioritized training of the most critical components. Second, we introduce an adaptive progressive schedule that adaptively increases computational complexity during training by measuring the convergence efficiency. Ent-Prog reduces both training time and GPU memory consumption while maintaining model performance. Extensive experiments across three datasets, demonstrate the effectiveness of Ent-Prog, achieving up to 2.2x training speedup and 2.4x GPU memory reduction without compromising generative performance.