Update index.html

Author: MaruyamaAya

docs/blog/index.html

@@ -211,7 +211,7 @@ <h1 class="title is-2 publication-title" style="font-size: 2.12rem">
     <div class="container is-max-desktop">
       <div class="content has-text-justified">
         <p>
-          Towards efficienct Diffusion Transformers! We introduce <b><span style="color: #39c7c3;">R</span>egion-<span style="color: #7f51cf;">A</span>daptive <span style="color: #7f51cf;">S</span>ampling</b>, the first diffusion sampling strategy that allows for regional variability in sampling ratios. Compared to spatially uniform samplers, this flexibility enables our approach to allocate DiT's processing power to the model's current areas of interest, significantly improving generation quality within the same inference budget. With models like Lumina-Next-T2I and Stable Diffusion 3, <b><span style="color: #39c7c3;">R</span><span style="color: #7f51cf;">A</span><span style="color: #7f51cf;">S</span></b>'s fast-region noise updating yields <b>over 2x</b> the acceleration with negligible image quality loss. A user study comparing our method to uniform sampling across various generated cases further shows that our method maintains comparable generation quality at <b>1.6x</b> the acceleration rate.
+          Towards efficient Diffusion Transformers! We introduce <b><span style="color: #39c7c3;">R</span>egion-<span style="color: #7f51cf;">A</span>daptive <span style="color: #7f51cf;">S</span>ampling</b>, the first diffusion sampling strategy that allows for regional variability in sampling ratios. Compared to spatially uniform samplers, this flexibility enables our approach to allocate DiT's processing power to the model's current areas of interest, significantly improving generation quality within the same inference budget. With models like Lumina-Next-T2I and Stable Diffusion 3, <b><span style="color: #39c7c3;">R</span><span style="color: #7f51cf;">A</span><span style="color: #7f51cf;">S</span></b>'s fast-region noise updating yields <b>over 2x</b> the acceleration with negligible image quality loss. A user study comparing our method to uniform sampling across various generated cases further shows that our method maintains comparable generation quality at <b>1.6x</b> the acceleration rate.
           <br>
         </p>
       </div>