<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Homes on</title><link>https://c-allergic.github.io/home/</link><description>Recent content in Homes on</description><generator>Hugo</generator><language>en</language><atom:link href="https://c-allergic.github.io/home/index.xml" rel="self" type="application/rss+xml"/><item><title>Home</title><link>https://c-allergic.github.io/home/home/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://c-allergic.github.io/home/home/</guid><description>&lt;section id="showcase" class="rad-showcase rad-showcase--index rad-animation-group rad-fade-down"&gt;
 &lt;div&gt;
 

 &lt;div class="showcase-section row flex-column-reverse flex-md-row rad-fade-down rad-waiting rad-animate"&gt;
 &lt;div class="col-12 col-sm-12 col-md-6 col-lg-6 col-xl-6"&gt;
 &lt;h1 class="display-1"&gt;
 Hi, I&amp;#39;m Xiaofeng Li (Sheldon)
 &lt;/h1&gt;
 &lt;h2 class="display-5"&gt;
 &lt;span&gt;HKUST(GZ) · Image Caption Evaluation · Generation Quality Evaluation&lt;/span&gt;
 &lt;/h2&gt;
 &lt;div class="lead"&gt;
 I research how to reliably evaluate generative models — from image caption metrics to VLM-as-judge approaches.
 &lt;/div&gt;
 
 &lt;a
 href="https://github.com/c-allergic"
 class="btn btn-primary"
 &gt;
 &lt;i class="icon-square-github"&gt;&lt;/i&gt;
 GitHub
 &lt;/a&gt;
 
 &lt;/div&gt;
 &lt;div class="col-12 col-sm-12 col-md-6 col-lg-6 col-xl-6 profile-image"&gt;
 










 
 



 
 


 &lt;picture class=""&gt;
 
 

 &lt;source srcset="https://c-allergic.github.io/images/showcase/showcase_hu_8ad1fbd4c9281872.webp 1x, https://c-allergic.github.io/images/showcase/showcase_hu_c000e9d626435f9b.webp 2x" type="image/webp" /&gt;
 &lt;source srcset="https://c-allergic.github.io/images/showcase/showcase_hu_534b425d380a36aa.png 1x, https://c-allergic.github.io/images/showcase/showcase_hu_d005f99e55f9a6b3.png 2x" type="image/png"&gt;

 &lt;img
 width="400"
 height="400"
 class="lozad img-responsive "
 data-src="https://c-allergic.github.io/images/showcase/showcase_hu_534b425d380a36aa.png"
 alt=''
 src="data:image/gif;base64,R0lGODlhBwACAIAAAP///wAAACH5BAEAAAEALAAAAAAHAAIAAAIDjI9YADs="
 loading="lazy"
 decoding="async"
 /&gt;
 &lt;/picture&gt;




 &lt;/div&gt;
 &lt;div class="col-12 inner-content"&gt;
 





&lt;div class="container"&gt;
 &lt;div class="row platform-links shortcode"&gt;
 &lt;div class="col-12"&gt;
 
 
 &lt;a
 href="https://github.com/c-allergic"
 target="_blank"
 rel="noopener noreferrer"
 aria-label="square-github"
 &gt;&lt;i class="icon-square-github"&gt;&lt;/i&gt;
 &lt;/a&gt;

 
 
 &lt;a
 href="mailto:xil683@connect.hkust-gz.edu.cn"
 target="_blank"
 rel="noopener noreferrer"
 aria-label="email"
 &gt;&lt;i class="icon-email"&gt;&lt;/i&gt;
 &lt;/a&gt;

 
 
 &lt;a
 href="https://arxiv.org/abs/2603.26154"
 target="_blank"
 rel="noopener noreferrer"
 aria-label="file-pdf"
 &gt;&lt;i class="icon-file-pdf"&gt;&lt;/i&gt;
 &lt;/a&gt;

 
 
 &lt;a
 href="https://github.com/c-allergic/awesome-generation-model-evaluation"
 target="_blank"
 rel="noopener noreferrer"
 aria-label="table-list"
 &gt;&lt;i class="icon-table-list"&gt;&lt;/i&gt;
 &lt;/a&gt;

&lt;/div&gt;
 &lt;/div&gt;
&lt;/div&gt;



 &lt;/div&gt;
 

 &lt;/div&gt;


 &lt;div class="row platform-links"&gt;
 &lt;div class="col-12"&gt;&lt;/div&gt;
 &lt;/div&gt;
 &lt;/div&gt;
 
&lt;/section&gt;
















&lt;section id="about" class="section rad-animation-group pb-0"&gt;
 &lt;div class="rad-fade-down"&gt;
 &lt;div class="row d-flex flex-column-reverse flex-md-row"&gt;
 &lt;div class="about__profile-picture col-12 col-md-6"&gt;
 










 
 



 
 


 &lt;picture class=""&gt;
 
 

 &lt;source srcset="https://c-allergic.github.io/images/about/user-picture_hu_8ad1fbd4c9281872.webp 1x, https://c-allergic.github.io/images/about/user-picture_hu_c000e9d626435f9b.webp 2x" type="image/webp" /&gt;
 &lt;source srcset="https://c-allergic.github.io/images/about/user-picture_hu_534b425d380a36aa.png 1x, https://c-allergic.github.io/images/about/user-picture_hu_d005f99e55f9a6b3.png 2x" type="image/png"&gt;

 &lt;img
 width="400"
 height="400"
 class="lozad img-responsive image-left-overflow"
 data-src="https://c-allergic.github.io/images/about/user-picture_hu_534b425d380a36aa.png"
 alt=''
 src="data:image/gif;base64,R0lGODlhBwACAIAAAP///wAAACH5BAEAAAEALAAAAAAHAAIAAAIDjI9YADs="
 loading="lazy"
 decoding="async"
 /&gt;
 &lt;/picture&gt;

 &lt;/div&gt;
 &lt;div class="col-12 col-md-6 my-auto"&gt;
 &lt;h2&gt;About Me&lt;/h2&gt;
 &lt;div class="about-me content"&gt;
 &lt;p&gt;I am a researcher at HKUST(GZ) working on &lt;strong&gt;evaluation of generative models&lt;/strong&gt;. My current focus is on &lt;strong&gt;Image Caption Evaluation&lt;/strong&gt; and &lt;strong&gt;Generation Quality Evaluation&lt;/strong&gt; — tackling how we can reliably and comprehensively measure what makes generated content good.&lt;/p&gt;</description></item></channel></rss>