Highlight metrics

README.md

@@ -37,6 +37,13 @@ This model was fine-tuned from [Stable Diffusion XL](https://huggingface.co/stab
 
 Refer to our code repository [here](https://github.com/mapo-t2i/mapo). 
 
+## Qualitative Comparison
+
+<div align="center">
+<img src="assets/comparison.png" width=750/>
+</div>
+
+
 ## Results
 
 Below we report some quantitative metrics and use them to compare MaPO to existing models: 
@@ -86,12 +93,12 @@ Below we report some quantitative metrics and use them to compare MaPO to existi
             <td>Diffusion-DPO</td>
             <td>6.03</td>
             <td>31.1</td>
-            <td>22.7</td>
+            <td><b>22.7<b/></td>
         </tr>
-        <tr class="ours">
-            <td>MaPO (Ours)</td>
-            <td>6.17</td>
-            <td>31.2</td>
+        <tr>
+            <td><b>MaPO (Ours)<b/></td>
+            <td><b>6.17<b/></td>
+            <td><b>31.2<b/></td>
             <td>22.5</td>
         </tr>
     </tbody>
@@ -119,17 +126,17 @@ The table below reports memory efficiency of MaPO, making it a better alternativ
         <tr>
             <td>Time (↓)</td>
             <td>63.5</td>
-            <td>54.3 (-14.5%)</td>
+            <td><b>54.3 (-14.5%)</b></td>
         </tr>
         <tr>
             <td>GPU Mem. (↓)</td>
             <td>55.9</td>
-            <td>46.1 (-17.5%)</td>
+            <td><b>46.1 (-17.5%)</b></td>
         </tr>
         <tr>
             <td>Max Batch (↑)</td>
             <td>4</td>
-            <td>16 (×4)</td>
+            <td><b>16 (×4)</b></td>
         </tr>
     </tbody>
 </table>