<ul class="dashed" data-apple-notes-indent-amount="0"><li><span style="font-family: '.PingFangUITextSC-Regular'">文章标题：</span>JUST-DUB-IT: Video Dubbing via Joint Audio-Visual Diffusion</li><li><span style="font-family: '.PingFangSC-Regular'">文章地址：</span><a href="https://arxiv.org/abs/2601.22143">https://arxiv.org/abs/2601.22143</a> </li><li>SIGGRAPH 2026</li></ul> <img src="https://imagedelivery.net/phxEHgsq3j8gSnfNAJVJSQ/node3_e3bc3a5e-61cf-43af-89f2-aab816de254b/public" style="background-color:initial;max-width:min(100%,1398px);max-height:min(1222px);;background-image:url(https://imagedelivery.net/phxEHgsq3j8gSnfNAJVJSQ/node3_e3bc3a5e-61cf-43af-89f2-aab816de254b/public);height:auto;width:100%;object-fit:cover;background-size:cover;display:block;" width="1398" height="1222"><ul class="dashed" data-apple-notes-indent-amount="0"><li></li></ul> <img src="https://imagedelivery.net/phxEHgsq3j8gSnfNAJVJSQ/node3_07314e7d-0c7d-4e37-b20a-91e4fbf46d47/public" style="background-color:initial;max-width:min(100%,1440px);max-height:min(724px);;background-image:url(https://imagedelivery.net/phxEHgsq3j8gSnfNAJVJSQ/node3_07314e7d-0c7d-4e37-b20a-91e4fbf46d47/public);height:auto;width:100%;object-fit:cover;background-size:cover;display:block;" width="1440" height="724"> 文章做的是视频编辑的任务，将一段视频的语言转成另一种语言，要求视频和音频同步编辑。文章利用音视频统一模型的强大先验能力进行数据的构造，如第二张图（这里的inpainting没太理解，文章里也没具体说明）。随后利用IC-LoRA的方式进行训练。