update

index.html

@@ -132,7 +132,29 @@ <h3 class="title is-4">NeurIPS 2024</h3>
                   <span class="icon">
                       <i class="fa fa-database"></i>
                   </span>
-                  <span>Model</span>
+                  <span>Models</span>
+                  </a>
+              </span>
+
+              <!-- Poster Link. -->
+              <span class="link-block">
+                <a target="_blank" href="https://manchery.github.io/assets/pub/nips2024_ivideogpt/poster.pdf"
+                  class="external-link button is-normal is-rounded is-dark">
+                  <span class="icon">
+                      <i class="fa fa-image"></i>
+                  </span>
+                  <span>Poster</span>
+                  </a>
+              </span>
+
+              <!-- Slides Link. -->
+              <span class="link-block">
+                <a target="_blank" href="https://manchery.github.io/assets/pub/nips2024_ivideogpt/slides.pdf"
+                  class="external-link button is-normal is-rounded is-dark">
+                  <span class="icon">
+                      <i class="fa fa-film"></i>
+                  </span>
+                  <span>Slides</span>
                   </a>
               </span>
             </div>
@@ -281,6 +303,62 @@ <h3 class="title is-4">Video Samples: Open X-Embodiment (Action-free)</h3>
         </div>
         <p class="content has-text-centered"><i>In each pair, the left image is the ground truth and the right image is the predicted video.
           The red border indicates the context frame and the green border means the frame was predicted. <b>Hover to zoom in.</b></i></p>
+      </div>
+      <br/>
+      <h3 class="title is-4">Open X-Embodiment (Goal-conditioned)</h3>
+      <p class="content has-text-justified">
+        A sequence of tokens provides a flexible way to specify tasks, inputs, and outputs. A variant of iVideoGPT for <b>goal-conditioned video prediction</b> 
+        is simply achieved by rearranging the frame sequence, while keeping the architecture and training procedure consistent.
+      </p>
+      <div class="zoom_container has-text-centered">
+        <div id="content_zoom_out">
+          <img src="./static/gifs/oxe_goal_cond_1.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_2.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_3.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_4.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_5.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_6.gif" style="width:16%"/>
+          <br/>
+          <img src="./static/gifs/oxe_goal_cond_7.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_8.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_9.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_10.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_11.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_12.gif" style="width:16%"/>
+          <br/>
+          <img src="./static/gifs/oxe_goal_cond_13.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_14.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_15.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_16.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_17.gif" style="width:16%"/>
+          <img src="./static/gifs/oxe_goal_cond_18.gif" style="width:16%"/>
+        </div>
+        <div id="content_zoom_in">
+          <img src="./static/gifs/oxe_goal_cond_1.gif" style="width:33%"/>
+          <img src="./static/gifs/oxe_goal_cond_2.gif" style="width:33%"/>
+          <img src="./static/gifs/oxe_goal_cond_3.gif" style="width:33%"/>
+          <br/>
+          <img src="./static/gifs/oxe_goal_cond_4.gif" style="width:33%"/>
+          <img src="./static/gifs/oxe_goal_cond_5.gif" style="width:33%"/>
+          <img src="./static/gifs/oxe_goal_cond_6.gif" style="width:33%"/>
+          <br/>
+          <img src="./static/gifs/oxe_goal_cond_7.gif" style="width:33%"/>
+          <img src="./static/gifs/oxe_goal_cond_8.gif" style="width:33%"/>
+          <img src="./static/gifs/oxe_goal_cond_9.gif" style="width:33%"/>
+          <br/>
+          <img src="./static/gifs/oxe_goal_cond_10.gif" style="width:33%"/>
+          <img src="./static/gifs/oxe_goal_cond_11.gif" style="width:33%"/>
+          <img src="./static/gifs/oxe_goal_cond_12.gif" style="width:33%"/>
+          <br/>
+          <img src="./static/gifs/oxe_goal_cond_13.gif" style="width:33%"/>
+          <img src="./static/gifs/oxe_goal_cond_14.gif" style="width:33%"/>
+          <img src="./static/gifs/oxe_goal_cond_15.gif" style="width:33%"/>
+          <br/>
+          <img src="./static/gifs/oxe_goal_cond_16.gif" style="width:33%"/>
+          <img src="./static/gifs/oxe_goal_cond_17.gif" style="width:33%"/>
+          <img src="./static/gifs/oxe_goal_cond_18.gif" style="width:33%"/>
+        </div>
+        <p class="content has-text-centered"><i>In each pair, the left image is the ground truth and the right image is the predicted video. <b>Hover to zoom in.</b></i></p>
       </div>      
       <br/>
       <details>