<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Sentiment Classification on Cairo Cananea</title>
    <link>https://cairocananea.com.br/en/tags/sentiment-classification/</link>
    <description>Recent content in Sentiment Classification on Cairo Cananea</description>
    <generator>Hugo -- 0.152.2</generator>
    <language>en</language>
    <lastBuildDate>Tue, 19 May 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://cairocananea.com.br/en/tags/sentiment-classification/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Beyond Full Fine-Tuning: LoRA with DistilBERT for Early Dissatisfaction Detection in Ecommerce Reviews</title>
      <link>https://cairocananea.com.br/en/projects/amazon-reviews-classification-lora/</link>
      <pubDate>Tue, 19 May 2026 00:00:00 +0000</pubDate>
      <guid>https://cairocananea.com.br/en/projects/amazon-reviews-classification-lora/</guid>
      <description>Beyond full fine-tuning: LoRA achieves 0.943 F1 and 0.996 AUC with just 1% of parameters. But silent bugs (wrong target modules, PEFT breaking Captum) and the conceptual misconception of LoRA as a mere redirector reveal the real trade-offs of the method.</description>
    </item>
  </channel>
</rss>
