VAD：多模态同策略蒸馏中的视觉证据归因目标重建

AI 摘要

本文提出视觉归因蒸馏（VAD），一种用于多模态同策略蒸馏的反事实目标重建算法，通过评估教师模型在有/无视觉证据时的输出差异，估计纠正信号中由视觉证据支持的部分。在4B和9B规模的六个细粒度视觉基准上，VAD优于直接特权视图蒸馏和视觉优势加权方法。结果表明，反事实目标重建是源混合监督的有效替代方案。

正文节选

VAD: Attributing Visual Evidence for Target Reconstruction in Multimodal On-Policy Distillation Abstract Multimodal on-policy distillation (OPD) transfers fine-grained visual knowledge by supervising student-generated trajectories with a privileged-view teacher. Yet its next-token corrections are source-mixed, combining visual signals with linguistic priors and teacher-specific effects. The key challenge is to estimate which corrections are supported by visual evidence, not merely where or how s

发布时间：—

抓取时间：2026-08-04 11:04

来源机构：Hugging Face