<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: thetruecopy</title>
    <description>The latest articles on DEV Community by thetruecopy (@thetruecopy).</description>
    <link>https://dev.to/thetruecopy</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4059280%2Fa18de296-c398-47a4-ad61-64599eaa7209.png</url>
      <title>DEV Community: thetruecopy</title>
      <link>https://dev.to/thetruecopy</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/thetruecopy"/>
    <language>en</language>
    <item>
      <title>Building an AI Text Detector with RoBERTa &amp; Burstiness Analysis</title>
      <dc:creator>thetruecopy</dc:creator>
      <pubDate>Sun, 02 Aug 2026 15:32:16 +0000</pubDate>
      <link>https://dev.to/thetruecopy/building-an-ai-text-detector-with-roberta-burstiness-analysis-2chk</link>
      <guid>https://dev.to/thetruecopy/building-an-ai-text-detector-with-roberta-burstiness-analysis-2chk</guid>
      <description>&lt;p&gt;Hey everyone! I just deployed &lt;strong&gt;Veridraft AI Detector Pro&lt;/strong&gt;, a Streamlit app designed to catch AI-generated text without relying solely on basic word probabilities.&lt;/p&gt;

&lt;h3&gt;
  
  
  Key Features
&lt;/h3&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Global Context Scan:&lt;/strong&gt; Evaluates overall document context alongside individual sentence windowing.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Burstiness Scoring:&lt;/strong&gt; Measures sentence length variance (&lt;code&gt;CV = std_dev / mean&lt;/code&gt;) to catch uniform, robotic structures.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Normalized Parsing:&lt;/strong&gt; Cleans hidden characters in PDF/DOCX uploads so outputs match raw text.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Tech Stack:&lt;/strong&gt; Python, Streamlit Cloud, Hugging Face (&lt;code&gt;Hello-SimpleAI/chatgpt-detector-roberta&lt;/code&gt;), PyTorch, PyPDF.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;🔗 &lt;strong&gt;Live Demo:&lt;/strong&gt; &lt;a href="https://thetruecopy-veridraft-detector-app-thguvb.streamlit.app" rel="noopener noreferrer"&gt;https://thetruecopy-veridraft-detector-app-thguvb.streamlit.app&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Would love for you to test it out and drop any feedback or edge cases!&lt;/p&gt;

</description>
      <category>python</category>
      <category>ai</category>
      <category>machinelearning</category>
      <category>streamlit</category>
    </item>
  </channel>
</rss>
