edit

index.html

@@ -165,6 +165,24 @@ <h1 class="title is-1 publication-title">LLAMA PRO: Progressive LLaMA with Block
 <section class="hero teaser">
   <div class="container is-max-desktop">
     <div class="hero-body">
+      <table width="77%" align="center" border="0" cellspacing="0" cellpadding="20"><tbody>
+        <tr>
+          <td>
+            <heading>🔥<b>What's New</b></heading>
+            <div style="height: 150px; overflow: auto;">
+              <ul> 
+                <li> <font color="red">[2024/02/23] We release MetaMath-Mistral-Pro that surpasses previous MetaMath series 7B models at both GSM8k and MATH. The evaluation is following the official MetaMath repo.</font>
+                <li> [2024/02/23] We release the evaluation code of Mistral-Pro-8B-v0.1 in lm-evaluation-harness.
+                <li> [2024/02/23] We release the Mistral-Pro-8B-v0.1 with superior performance on a range of benchmarks. It enhances the code and math performance of Mistral and matches the performance of the recently dominant model, Gemma. assets/mistral_pro_performance.png
+                <li> [2024/01/18] Add the training code in open-instruct.
+                <li> [2024/01/07] Add how to run gradio demo locally in demo
+                <li> [2024/01/06] We open source the LLaMA-Pro repository and Demo & Model.
+              </ul>
+            </div>
+          </td>
+        </tr>
+      </tbody></table>
+      <br>
       <h2 class="subtitle has-text-centered">
         An overview of <span class="dnerf">LLaMA-Pro</span> of two stages: <br>
       </h2>