diff --git a/Toolformer-Language-Models-Can-Teach-Themselves-to-Use-Tools.html b/Toolformer-Language-Models-Can-Teach-Themselves-to-Use-Tools.html
index d34cb91..ee172aa 100644
--- a/Toolformer-Language-Models-Can-Teach-Themselves-to-Use-Tools.html
+++ b/Toolformer-Language-Models-Can-Teach-Themselves-to-Use-Tools.html
@@ -200,7 +200,7 @@ <h2 id="approach">Approach</h2>
     <p>Starting with a language model, M, the goal is to enable the language model to use tools by invoking API calls.</p>
   </li>
   <li>
-    <p>An API call is denoted by the tuple $c = (api-name, api-input)$. It can be linearized as $e(c) = [api-name(api-input)]$ or as $e(c, r) = [api-name(api-input) -&gt; r]$ where $r$ denotes the result of the API.</p>
+    <p>An API call is denoted by the tuple $c =$ (api_name, api_input). It can be linearized as $e(c) =$ [api_name(api_input)$]$ or as $e(c, r) = [$api_name(api_input) $ -&gt; r]$ where $r$ denotes the result of the API.</p>
   </li>
   <li>
     <p>The given dataset of plain text, $C$, is converted into a dataset $C*$ augmented with the API calls using a three-step process.</p>
diff --git a/atom.xml b/atom.xml
index a68100d..26ef666 100644
--- a/atom.xml
+++ b/atom.xml
@@ -4,7 +4,7 @@
  <title>Papers I Read</title>
  <link href="https://shagunsodhani.github.io/papers-I-read/atom.xml" rel="self"/>
  <link href="https://shagunsodhani.github.io/papers-I-read/"/>
- <updated>2023-02-12T14:01:15-05:00</updated>
+ <updated>2023-02-12T14:08:58-05:00</updated>
  <id>https://shagunsodhani.github.io/papers-I-read</id>
  <author>
    <name>Shagun Sodhani</name>
@@ -35,7 +35,7 @@
     &lt;p&gt;Starting with a language model, M, the goal is to enable the language model to use tools by invoking API calls.&lt;/p&gt;
   &lt;/li&gt;
   &lt;li&gt;
-    &lt;p&gt;An API call is denoted by the tuple $c = (api-name, api-input)$. It can be linearized as $e(c) = [api-name(api-input)]$ or as $e(c, r) = [api-name(api-input) -&amp;gt; r]$ where $r$ denotes the result of the API.&lt;/p&gt;
+    &lt;p&gt;An API call is denoted by the tuple $c =$ (api_name, api_input). It can be linearized as $e(c) =$ [api_name(api_input)$]$ or as $e(c, r) = [$api_name(api_input) $ -&amp;gt; r]$ where $r$ denotes the result of the API.&lt;/p&gt;
   &lt;/li&gt;
   &lt;li&gt;
     &lt;p&gt;The given dataset of plain text, $C$, is converted into a dataset $C*$ augmented with the API calls using a three-step process.&lt;/p&gt;