quote

index.html

@@ -443,8 +443,14 @@ <h2>Overview</h2>
 
         <p>
             We identify that the main challenge in training diffusion models stems from the need to learn a high-quality internal representation.
-            In particular, we show the performance of <b>generative diffusion models can be improved dramatically when they are supported by an
-            external representation</b>—a high-quality, pre-trained representation from another model, such as a self-supervised visual encoder.
+            In particular, we show: 
+        </p>
+        <p>
+            <q>    
+                <b>The performance of generative diffusion models can be improved dramatically when they are supported by an
+                external representation—a high-quality, pre-trained representation from another model, such as a self-supervised visual encoder.
+            </b>
+            </q>    
         </p>
         <p>
             Specifically, we introduce <b>REPresentation Alignment (REPA)</b>, a simple regularization technique