跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.23025 · v1

When a Larger Guidance Scale Repairs Text-to-Video Failures and When It Does Not

首次提交 2026-11-18 00:30
版本更新 2026-11-18 00:30
雷达观测 2026-11-18 16:38
arXiv ↗PDF ↗

摘要

When a text-to-video diffusion model ignores part of a prompt, the usual first fix is to raise the classifier-free guidance scale. This paper is a measurement study of when that fix works. On Mochi, the same increase repairs failures on rare compositions ("a snowy beach") beyond what generating again does, does not repair negations ("a highway with no vehicles"), where it fixes fewer failures than generating again does, and changes nothing for events that must happen late in a clip ("a sandcastle that a wave washes flat"). Which of these happens can be known in advance, from the prompt and one inexpensive score. The score, which we call demand, measures how strongly the denoiser reacts to the part of the prompt that the failed video leaves out, and it is read from the first quarter of denoising, before any video exists. Our main finding is a condition: raising the scale is effective only when the prompt contains no negation and no event tied to a specific time, and when demand is sufficiently high, with recovery clearly higher above a fitted threshold than below it. Published corrections for the three failure types, used as fixed probes, behave as this account predicts. A short decision rule follows: before generating again, a user can tell whether the most common fix is worth trying, and what to use when it is not.

作者备注:29 pages, 15 figures

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-18 00:30 · 最新观测版本

When a Larger Guidance Scale Repairs Text-to-Video Failures and When It Does Not

When a text-to-video diffusion model ignores part of a prompt, the usual first fix is to raise the classifier-free guidance scale. This paper is a measurement study of when that fix works. On Mochi, the same increase repairs failures on rare compositions ("a snowy beach") beyond what generating again does, does not repair negations ("a highway with no vehicles"), where it fixes fewer failures than generating again does, and changes nothing for events that must happen late in a clip ("a sandcastle that a wave washes flat"). Which of these happens can be known in advance, from the prompt and one inexpensive score. The score, which we call demand, measures how strongly the denoiser reacts to the part of the prompt that the failed video leaves out, and it is read from the first quarter of denoising, before any video exists. Our main finding is a condition: raising the scale is effective only when the prompt contains no negation and no event tied to a specific time, and when demand is sufficiently high, with recovery clearly higher above a fitted threshold than below it. Published corrections for the three failure types, used as fixed probes, behave as this account predicts. A short decision rule follows: before generating again, a user can tell whether the most common fix is worth trying, and what to use when it is not.

查看此版本解读 →查看此版本原文 ↗