Update index.html

index.html

@@ -95,6 +95,13 @@ <h1>FRACTURED-SORRY-Bench</h1>
         <p class="subtitle">Framework for Revealing Attacks in Conversational Turns Undermining Refusal Efficacy and Defenses over SORRY-Bench</p>
     </header>
 
+    <div class="section">
+        <h2>Get Started</h2>
+        <a href="https://github.com/AmanPriyanshu/FRACTURED-SORRY-Bench/" class="button">View on GitHub</a>
+        <a href="https://huggingface.co/datasets/AmanPriyanshu/FRACTURED-SORRY-Bench" class="button">Check out the Dataset</a>
+        <a href="https://amanpriyanshu.github.io/FRACTURED-SORRY-Bench/FRACTURED_SORRY_Bench.pdf" class="button">Read the Paper</a>
+    </div>
+
     <div class="section">
         <h2>About FRACTURED-SORRY-Bench</h2>
         <p>FRACTURED-SORRY-Bench is a framework for evaluating the safety of Large Language Models (LLMs) against <span class="highlight">multi-turn conversational attacks</span>. Building upon the SORRY-Bench dataset, we propose a simple yet effective method for generating adversarial prompts by breaking down harmful queries into seemingly innocuous sub-questions.</p>
@@ -206,8 +213,9 @@ <h3>Decomposed Responses</h3>
     </div>
 
     <div class="section">
-        <h2>Get Started</h2>
+        <h2>Important Links</h2>
         <a href="https://github.com/AmanPriyanshu/FRACTURED-SORRY-Bench/" class="button">View on GitHub</a>
+        <a href="https://huggingface.co/datasets/AmanPriyanshu/FRACTURED-SORRY-Bench" class="button">Check out the Dataset</a>
         <a href="https://amanpriyanshu.github.io/FRACTURED-SORRY-Bench/FRACTURED_SORRY_Bench.pdf" class="button">Read the Paper</a>
     </div>