<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
	xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
		>

<channel>
	<title>UniResearch Community | xingxing | Activity</title>
	<link>https://community.uniresearch.ai/members/admin/activity/</link>
	<atom:link href="https://community.uniresearch.ai/members/admin/activity/feed/" rel="self" type="application/rss+xml" />
	<description>Activity feed for xingxing.</description>
	<lastBuildDate>Fri, 11 Sep 2026 11:08:05 +0800</lastBuildDate>
	<generator>https://buddypress.org/?v=3.4.4</generator>
	<language>en-US</language>
	<ttl>30</ttl>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>2</sy:updateFrequency>
		
								<item>
				<guid isPermaLink="false">a802f20e3a08c455236389b9fc74c151</guid>
				<title>xingxing posted a new activity comment</title>
				<link>https://community.uniresearch.ai/news-feed/p/158/#acomment-324</link>
				<pubDate>Thu, 20 Aug 2026 06:39:18 +0800</pubDate>

									<content:encoded><![CDATA[<p>This really hits home. It’s always when you stop chasing happiness that it quietly slips in through the back door.</p>
				<strong>In reply to</strong> -
					<a href="https://community.uniresearch.ai/members/bb-trent/" data-bb-hp-profile="10" rel="nofollow">Trent</a> posted an update in the group <a href="https://community.uniresearch.ai/groups/llm-practices-for-scientific-research/" data-bb-hp-group="5" rel="nofollow">LLM Practices for Scientific Research</a> Happiness often sneaks in through a door you didn&#8217;t know you left open.					]]></content:encoded>
				
				
							</item>
					<item>
				<guid isPermaLink="false">dc51bc7430a5762d035cdc3e2ead27e6</guid>
				<title>xingxing posted an update: Just finished a deep dive into the sparse attention [&#133;]</title>
				<link>https://community.uniresearch.ai/news-feed/p/319/</link>
				<pubDate>Mon, 17 Aug 2026 03:50:33 +0800</pubDate>

									<content:encoded><![CDATA[<p>Just finished a deep dive into the sparse attention variants from the &#8220;Longformer&#8221; paper. The implementation seems promising for long sequences, but I&#8217;m wondering if it&#8217;s practical for low-resource NLP tasks. Has anyone experimented with it on small datasets? Curious about your experience.</p>
]]></content:encoded>
				
									<slash:comments>0</slash:comments>
				
							</item>
		
	</channel>
</rss>
		