本文深入探讨了离散扩散模型的理论基础，通过分析其目标函数来探索它们学习到的内容。研究引入了“Oracle Distance”定理，该定理认为负ELBO精确等于数据熵加上Oracle反向过程与学习到的反向过程之间的路径KL散度。该框架允许在去噪器、空腔预测器和分数函数等不同参数化之间进行精确转换，并解释了在各种扩散模型实现中观察到的差异。 AI
影响 为理解和优化离散扩散模型提供了理论框架，可能带来更高效、更有效的生成式AI。
排序理由 详细介绍离散扩散模型理论进展的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2607.05381v1 Announce Type: cross Abstract: What does a discrete diffusion model learn: a denoiser, a score ratio, or a bridge plug-in predictor? At the level of jump rates, these are one object in different coordinates, and reading a neural network in the wrong coordinate …
What does a discrete diffusion model learn: a denoiser, a score ratio, or a bridge plug-in predictor? At the level of jump rates, these are one object in different coordinates, and reading a neural network in the wrong coordinate changes the process being trained and sampled. Sta…