<?xml version="1.0" standalone="yes"?>
<Paper uid="W00-0409">
  <Title>Multi-document Summarization by Visualizing Topical Content</Title>
  <Section position="11" start_page="95" end_page="96" type="concl">
    <SectionTitle>
9 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have described methods for leveraging automatic summarization in the automatic generation of multimedia briefings. This work has taken, an open-domain approach, in order to meet the requirements of the DARPA application we are involved with. We believe there is a stronger role that NL generation can play in the narrative aspects of our briefings, which currently rely for the most part on canned text. Our future work on description merging in biographical summaries, and on introducing referring expressions into the narrative nodes, would in effect take advantage of more powerful generation methods, without sacrificing open-domain capabilities. This may require much richer recta-information specifications than the ones we currently use.</Paragraph>
    <Paragraph position="1"> Finally, we have begun the design of the Script Creator GUI (the only component in Figure 1 remaining to be built). This will allow the author to create scripts for the briefing generator (instead of editing templates by hand), by laying out icons for media objects in temporal order. A user will be able to select a &amp;quot;standard&amp;quot; briefing template from a menu, and then view it in a briefing/template structure editor. The user can then provide content by adding annotations to any node in the briefing template. The user will have a choice of saving the edit version in template form, or in SMIL or possibly Microsoft Powerpoint format.</Paragraph>
  </Section>
class="xml-element"></Paper>