.

Author: faikit

blog/agibot-world/index.html

@@ -303,12 +303,14 @@ <h4 class="text-white w-full max-w-wide leading-loose ">
                         Does AgiBot World boost policy learning at scale?
                     </h4>
 
-        
-                    
-                    <figure class="w-full max-w-wide mt-3 mb-3 laptop:mt-6 laptop:mb-6">
+    
+
+                    <figure class="w-full max-w-wide mt-3 laptop:mt-0">
                         <div id="fig3-1"></div>
+                    </figure>
+                    <figure class="w-full max-w-wide -mt-3 mb-3 laptop:mb-6">
                         <div id="fig3-2"></div>
-                        <figcaption class="text-white text-xs mt-3 leading-loose">
+                        <figcaption class="text-white text-xs mt-3 laptop:mt-0 leading-loose">
                             Figure 3: Policies pre-trained on our dataset outperform those trained on OXE in both seen (0.77
                             v.s. 0.47) and out-of-distribution scenarios (0.67 v.s. 0.38).
                         </figcaption>
@@ -328,9 +330,9 @@ <h4 class="text-white w-full max-w-wide leading-loose ">
 
 
 
-                    <figure class="w-full max-w-wide mt-3 mb-3 laptop:mt-6 laptop:mb-6">
+                    <figure class="w-full max-w-wide mt-3 mb-3 laptop:mt-0 laptop:mb-6">
                         <div id="fig4"></div>
-                        <figcaption class="text-white text-xs mt-3 leading-loose">
+                        <figcaption class="text-white text-xs mt-3 laptop:mt-0 leading-loose">
                             Figure 4: We evaluate different models pre-trained on the AgiBot World dataset, where GO-1 significantly outperforms previous SOTA policy RDT.
                         </figcaption>
                     </figure>
@@ -349,12 +351,7 @@ <h4 class="text-white w-full max-w-wide leading-loose ">
 
 
 
-                    <figure class="w-full max-w-wide mt-3 mb-3 laptop:mt-6 laptop:mb-6">
-                        <div id="fig5"></div>
-                        <figcaption class="text-white text-xs mt-3 leading-loose">
-                            Figure 5: We evaluate how model performance scales with data size.
-                        </figcaption>
-                    </figure>
+