Updated on 2024-12-31

Author: lxaw

index.html

@@ -35,7 +35,7 @@ <h1>Where?</h1>
             <!-- Last Update Section -->
             <section>
                 <h1>When?</h1>
-        Last time this was edited was 2024-12-30 (YYYY/MM/DD).
+        Last time this was edited was 2024-12-31 (YYYY/MM/DD).
             </section>
 
             <!-- Footer -->

papers/list.json

@@ -1,11 +1,20 @@
 [
+  {
+    "title": "Prefix-Tuning: Optimizing Continuous Prompts for Generation",
+    "author": "Xiang Lisa Li et al",
+    "year": "2021",
+    "topic": "prefix-tuning, prompting, llm",
+    "venue": "Arxiv",
+    "description": "This paper proposes prefix-tuning, which keeps language model params frozen but optimizes a continuous task-specific vector (prefix).",
+    "link": "https://arxiv.org/pdf/2101.00190"
+  },
   {
     "title": "The Power of Scale for Parameter-Efficient Prompt Tuning",
     "author": "Brian Lester et al",
     "year": "2021",
     "topic": "prompting, llm",
     "venue": "Arxiv",
-    "description": "This paper explores adding soft prompts to condition forzen language models. Basically, soft prompts are learned through back-propagation and can be used to finetune language models without fully retraining. They also introduce the idea of \"prompt ensembling\" which is basically using multiple soft prompts on a model and ensembling their outputs.",
+    "description": "This paper explores adding soft prompts to condition frozen language models. Basically, soft prompts are learned through back-propagation and can be used to finetune language models without fully retraining. They also introduce the idea of \"prompt ensembling\" which is basically using multiple soft prompts on a model and ensembling their outputs.",
     "link": "https://arxiv.org/pdf/2104.08691"
   },
   {

papers_read.html

@@ -16,10 +16,10 @@ <h1>Here's where I keep a list of papers I have read.</h1>
         I typically use this to organize papers I found interesting. Please feel free to do whatever you want with it. Note that this is not every single paper I have ever read, just a collection of ones that I remember to put down.
     </p>
     <p id="paperCount">
-        So far, we have read 198 papers. Let's keep it up!
+        So far, we have read 199 papers. Let's keep it up!
     </p> 
     <small id="searchCount">
-        Your search returned 198 papers. Nice! 
+        Your search returned 199 papers. Nice! 
     </small>
 
     <div class="search-inputs">
@@ -46,13 +46,23 @@ <h1>Here's where I keep a list of papers I have read.</h1>
         </thead>
         <tbody>
 
+            <tr>
+                <td>Prefix-Tuning: Optimizing Continuous Prompts for Generation</td>
+                <td>Xiang Lisa Li et al</td>
+                <td>2021</td>
+                <td>prefix-tuning, prompting, llm</td>
+                <td>Arxiv</td>
+                <td>This paper proposes prefix-tuning, which keeps language model params frozen but optimizes a continuous task-specific vector (prefix).</td>
+                <td><a href="https://arxiv.org/pdf/2101.00190" target="_blank">Link</a></td>
+            </tr>
+        
             <tr>
                 <td>The Power of Scale for Parameter-Efficient Prompt Tuning</td>
                 <td>Brian Lester et al</td>
                 <td>2021</td>
                 <td>prompting, llm</td>
                 <td>Arxiv</td>
-                <td>This paper explores adding soft prompts to condition forzen language models. Basically, soft prompts are learned through back-propagation and can be used to finetune language models without fully retraining. They also introduce the idea of &quot;prompt ensembling&quot; which is basically using multiple soft prompts on a model and ensembling their outputs.</td>
+                <td>This paper explores adding soft prompts to condition frozen language models. Basically, soft prompts are learned through back-propagation and can be used to finetune language models without fully retraining. They also introduce the idea of &quot;prompt ensembling&quot; which is basically using multiple soft prompts on a model and ensembling their outputs.</td>
                 <td><a href="https://arxiv.org/pdf/2104.08691" target="_blank">Link</a></td>
             </tr>