how to generate feed entry for nonlatin document

[linusboyle]

Hi there, I'd like to use this repo as a starting point for my blog, part of which is in Chinese. However the title and summary of these documents in the Atom feed generated are garbled like this:

<entry>
    <title><#9E3F><#722A><#30FB>2021<#5E74>8<#6708></title>
    <link rel="alternate" type="text/html" hreflang="en" href="http://texmacs.github.io/notes/docs/record-2021-08.html"/>
    <id>texmacs.github.io/notes/record-2021-08.tm:2021-12-26T12:23:48Z</id>
    <updated>2021-12-26T12:23:48Z</updated>
    <published>2021-12-26T12:23:48Z</published>
    <summary><document><#300A><#4E2D><#56FD><#8BDD><#300B></document></summary>
  </entry>

Not so familiar with lisp, so I'm wondering if there's a function that can decode these.