--

Author: asudeh

efficient-llm-inference/index.html

@@ -24,7 +24,7 @@
           <div class="eyebrow">Project Website</div>
           <h1>Efficient LLM Inference</h1>
           <p class="subtitle">
-            Efficient (and provacy-preserving) inference on personal devices
+            Efficient (and privacy-preserving) inference on personal devices
           </p>
           <p class="hero-text">
             This project advances systems and algorithmic foundations for running large language models efficiently on personal devices, reducing dependence on cloud-only deployment while improving privacy, accessibility, and cost efficiency.
@@ -65,9 +65,6 @@ <h2>Project overview</h2>
           <p>
             This project aims to shift LLM inference toward a more user-centric and widely accessible model by enabling accurate, private, and efficient execution on personal computers. The broader goal is to support LLM operation without reliance on costly, energy-intensive accelerators or continuous internet connectivity, thereby expanding access to advanced language technologies.
           </p>
-          <p>
-            The project will deliver a unified framework for privacy-preserving on-device LLM inference. By improving efficiency on personal devices, this work seeks to reduce dependence on centralized infrastructure, strengthen data protection, and support more trustworthy and practical deployment of modern AI systems.
-          </p>
         </div>
       </div>
     </section>