Modified srank

Author: quochungto

srank-coderanker/index.html

@@ -390,13 +390,11 @@ <h2 class="title is-3">Conclusion</h2>
             We propose SRank, a novel reranking strategy designed to extract optimal code generation solutions from CodeLLMs.
             By modeling the relationships between clusters of code solutions, 
             we can more effectively identify the best solutions and improve the overall accuracy of code generation. 
-            We showcase the state-of-the-art performance of SRank on pass@1 across various well-known CodeLLMs, 
-            surpassing other ranking methods like CodeT and Coder-Reviewer in extensive evaluations.
           </p>
           <p>
-            We believe that SRank will be a valuable tool for developers, researchers, 
-            and anyone who uses code generation in their work. 
-            We are excited to see how this technology will be used to improve the way we write and interact with code.
+            We showcase the state-of-the-art performance of SRank on pass@1 across various well-known CodeLLMs, 
+            surpassing other ranking methods like CodeT and Coder-Reviewer in extensive evaluations.
+            Our findings also suggest that SRank can potentially address the challenges of code generation in real-world applications, illuminating strategies for selecting superior solutions within constrained coding environments.
           </p>
         </div>
       </div>