<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:wfw="http://wellformedweb.org/CommentAPI/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	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>paged attention &#8211; BIOENGINEER.ORG</title>
	<atom:link href="https://bioengineer.org/tag/paged-attention/feed/" rel="self" type="application/rss+xml" />
	<link>https://bioengineer.org</link>
	<description>Bioengineering</description>
	<lastBuildDate>Sun, 18 Oct 2026 05:59:52 +0000</lastBuildDate>
	<language>en-US</language>
	<sy:updatePeriod>
	hourly	</sy:updatePeriod>
	<sy:updateFrequency>
	1	</sy:updateFrequency>
	<generator>https://wordpress.org/?v=7.1.3</generator>

<image>
	<url>https://bioengineer.org/wp-content/uploads/2019/09/cropped-bioengineering-32x32.png</url>
	<title>paged attention &#8211; BIOENGINEER.ORG</title>
	<link>https://bioengineer.org</link>
	<width>32</width>
	<height>32</height>
</image> 
<site xmlns="com-wordpress:feed-additions:1">72741379</site>	<item>
		<title>New Co-Design Framework Speeds Up LLM Inference on Mixed Hardware by Taming Memory Bottlenecks</title>
		<link>https://bioengineer.org/new-co-design-framework-speeds-up-llm-inference-on-mixed-hardware-by-taming-memory-bottlenecks/</link>
		
		<dc:creator><![CDATA[]]></dc:creator>
		<pubDate>Sun, 18 Oct 2026 05:59:52 +0000</pubDate>
				<category><![CDATA[Technology]]></category>
		<category><![CDATA[algorithm-hardware co-design]]></category>
		<category><![CDATA[FP8 quantization]]></category>
		<category><![CDATA[grouped-query attention]]></category>
		<category><![CDATA[heterogeneous platforms]]></category>
		<category><![CDATA[KV cache optimization]]></category>
		<category><![CDATA[Large Language Models]]></category>
		<category><![CDATA[Llama]]></category>
		<category><![CDATA[LLM inference]]></category>
		<category><![CDATA[memory bandwidth]]></category>
		<category><![CDATA[paged attention]]></category>
		<category><![CDATA[throughput optimization]]></category>
		<category><![CDATA[vLLM]]></category>
		<guid isPermaLink="false">https://bioengineer.org/?p=433947</guid>

					<description><![CDATA[Researchers have unveiled LLM-CoOpt, an algorithm-hardware co-design framework that boosts large language model inference throughput by up to 13.43 percent and cuts latency by up to 16.79 percent through key-value cache optimization, grouped-query attention, and paged attention for long sequences.]]></description>
		
		
		
		<post-id xmlns="com-wordpress:feed-additions:1">433947</post-id>	</item>
	</channel>
</rss>
