Lost in Translation: Latent Concept Misalignment in Text-to-Image Diffusion Models

<section class="main">
    <div class="text-center-pub">
        <h2>Publications</h2>
        <p>&nbsp;</p>
        <p class="bold"><a href="https://lcmis.github.io"></a></p>
        <p><span class="bold">Juntu Zhao</span>, Junyu Deng<sup>*</sup>, Yixin Ye<sup>*</sup>, Chongxuan Li, Zhijie Deng<sup>†</sup>, Dequan Wang<sup>†</sup></p>
        <p class="italic">European Conference on Computer Vision (ECCV), 2024</p>
    </div>
</section>