<?xml version="1.0" encoding="utf-8" standalone="yes" ?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Multimedia Retrieval on DASYA</title>
    <link>https://dasya.itu.dk/tags/multimedia-retrieval/</link>
    <description>Recent content in Multimedia Retrieval on DASYA</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>en-us</language>
    <copyright>© DASYA, IT University of Copenhagen</copyright>
    <lastBuildDate>Fri, 21 Nov 2025 00:00:00 +0000</lastBuildDate>
    
	<atom:link href="https://dasya.itu.dk/tags/multimedia-retrieval/index.xml" rel="self" type="application/rss+xml" />
    
    
    <item>
      <title>Answering Questions from Multimedia Collections</title>
      <link>https://dasya.itu.dk/for-students/proposals/archive/osk-qa-multimedia/</link>
      <pubDate>Fri, 21 Nov 2025 00:00:00 +0000</pubDate>
      
      <guid>https://dasya.itu.dk/for-students/proposals/archive/osk-qa-multimedia/</guid>
      <description>Constrastive learning models have made it easier to find relevant content from multimedia collections through descriptive text queries, reducing the interactivity needed to solve simple tasks. However, for complex tasks not only pertaining visual elements, or tasks focused on answering questions involving the contents of a singular or multiple media items (videos/images) still requires a fair amount of interactivity.
 Retrieve results from query Inspect content (summaries, watching, browsing) Refine the query / add more context / different search strategy Repeat until answer to the question or relevant content has been discovered  RAG (Retrieval-Augmented-Generation) approaches utilize LLMs (Large-Language Models) to answer questions from a set of documents.</description>
    </item>
    
  </channel>
</rss>