deploy: f1178d37f5

2026-01-03 06:28:22 +00:00
parent 346f1f1450
commit 41ec0626e2
29 changed files with 32 additions and 32 deletions
--- a/posts/benchmarking-llms-on-jetson-orin-nano/index.html
+++ b/posts/benchmarking-llms-on-jetson-orin-nano/index.html
@@ -60,6 +60,6 @@ After running 66 inference tests across seven different language models ranging
 <a class=heading-link href=#references><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h2><ol><li><p>Williams, S., Waterman, A., & Patterson, D. (2009). &ldquo;Roofline: An Insightful Visual Performance Model for Multicore Architectures.&rdquo; <em>Communications of the ACM</em>, 52(4), 65-76.</p></li><li><p>NVIDIA Corporation. (2024). &ldquo;Jetson Orin Nano Developer Kit Technical Specifications.&rdquo; <a href=https://developer.nvidia.com/embedded/jetson-orin-nano-developer-kit class=external-link target=_blank rel=noopener>https://developer.nvidia.com/embedded/jetson-orin-nano-developer-kit</a></p></li><li><p>&ldquo;Jetson AI Lab Benchmarks.&rdquo; NVIDIA Jetson AI Lab. <a href=https://www.jetson-ai-lab.com/benchmarks.html class=external-link target=_blank rel=noopener>https://www.jetson-ai-lab.com/benchmarks.html</a></p></li><li><p>Gerganov, G., et al. (2023). &ldquo;GGML - AI at the edge.&rdquo; <em>GitHub</em>. <a href=https://github.com/ggerganov/ggml class=external-link target=_blank rel=noopener>https://github.com/ggerganov/ggml</a></p></li><li><p>Kwon, W., et al. (2023). &ldquo;Efficient Memory Management for Large Language Model Serving with PagedAttention.&rdquo; <em>Proceedings of SOSP 2023</em>.</p></li><li><p>Team, G., Mesnard, T., et al. (2025). &ldquo;Gemma 3: Technical Report.&rdquo; <em>arXiv preprint arXiv:2503.19786v1</em>. <a href=https://arxiv.org/html/2503.19786v1 class=external-link target=_blank rel=noopener>https://arxiv.org/html/2503.19786v1</a></p></li><li><p>Yang, A., et al. (2025). &ldquo;Qwen3 Technical Report.&rdquo; <em>arXiv preprint arXiv:2505.09388</em>. <a href=https://arxiv.org/pdf/2505.09388 class=external-link target=_blank rel=noopener>https://arxiv.org/pdf/2505.09388</a></p></li><li><p>DeepSeek-AI. (2025). &ldquo;DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning.&rdquo; <em>arXiv preprint arXiv:2501.12948v1</em>. <a href=https://arxiv.org/html/2501.12948v1 class=external-link target=_blank rel=noopener>https://arxiv.org/html/2501.12948v1</a></p></li><li><p>&ldquo;Running LLMs with TensorRT-LLM on NVIDIA Jetson Orin Nano Super.&rdquo; Collabnix. <a href=https://collabnix.com/running-llms-with-tensorrt-llm-on-nvidia-jetson-orin-nano-super/ class=external-link target=_blank rel=noopener>https://collabnix.com/running-llms-with-tensorrt-llm-on-nvidia-jetson-orin-nano-super/</a></p></li><li><p>Pope, R., et al. (2022). &ldquo;Efficiently Scaling Transformer Inference.&rdquo; <em>Proceedings of MLSys 2022</em>.</p></li><li><p>Frantar, E., et al. (2023). &ldquo;GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers.&rdquo; <em>Proceedings of ICLR 2023</em>.</p></li><li><p>Dettmers, T., et al. (2023). &ldquo;QLoRA: Efficient Finetuning of Quantized LLMs.&rdquo; <em>Proceedings of NeurIPS 2023</em>.</p></li><li><p>Lin, J., et al. (2023). &ldquo;AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration.&rdquo; <em>arXiv preprint arXiv:2306.00978</em>.</p></li></ol></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/breville-barista-pro-maintenance/index.html
+++ b/posts/breville-barista-pro-maintenance/index.html
@@ -23,6 +23,6 @@ Understanding the Two Primary Maintenance Cycles Link to heading The Breville Ba
 <a class=heading-link href=#routine-and-preventative-maintenance-schedule><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h4><p>In addition to the automated cycles, regular manual cleaning is essential for machine health.</p><p><strong>Daily Tasks:</strong></p><ul><li><strong>Purge Group Head:</strong> After the final use of the day, run hot water through the group head (without the portafilter) to clear grounds.</li><li><strong>Clean Portafilter & Baskets:</strong> Do not let used coffee grounds sit in the portafilter. Rinse with hot water after every use.</li><li><strong>Clean Steam Wand:</strong> Immediately after texturing milk, wipe the wand with a damp cloth and purge steam for 2-3 seconds to clear internal passages.</li><li><strong>Empty Drip Tray:</strong> Empty and rinse the drip tray regularly.</li></ul><p><strong>Weekly Tasks:</strong></p><ul><li><strong>Soak Components:</strong> Remove the filter basket from the portafilter. Soak both components in a solution of hot water and a cleaning tablet (or specific espresso cleaner) for 20-30 minutes to dissolve accumulated coffee oils. Rinse thoroughly.</li><li><strong>Clean Grinder:</strong> Empty the bean hopper. Run the grinder to clear any remaining beans, then use a brush and/or vacuum to clean out fines and oil residue from the burrs and chute.</li></ul><p><strong>Periodic Tasks (Every 2-3 Months):</strong></p><ul><li><strong>Replace Water Filter:</strong> The water filter located inside the water tank should be replaced every 3 months. This reduces the rate of scale buildup.</li><li><strong>Inspect Shower Screen:</strong> Use a brush to gently scrub the shower screen inside the group head to remove any stubborn coffee grounds.</li></ul><p>By adhering to this comprehensive maintenance schedule, you can ensure your Breville Barista Pro operates at peak performance and consistently produces high-quality espresso.</p><hr><p><strong>Reference:</strong></p><ul><li>Breville Barista Pro Instruction Manual and official manufacturer guidelines.</li></ul></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/espresso-theory-application-a-guide-for-the-breville-barista-pro/index.html
+++ b/posts/espresso-theory-application-a-guide-for-the-breville-barista-pro/index.html
@@ -18,6 +18,6 @@ Our overarching philosophy is simple: isolate and change only one variable at a
 <a class=heading-link href=#the-complete-dialing-in-workflow><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h3><p>This systematic process will get you to a delicious shot from your Breville Barista Pro efficiently:</p><ol><li><strong>Set Your Constants:</strong><ul><li><strong>Dose:</strong> <strong>18g</strong>.</li><li><strong>Ratio:</strong> <strong>1:2</strong> (meaning a <strong>Yield</strong> of <strong>36g</strong>).</li><li><strong>Pre-infusion:</strong> Use a consistent method (e.g., manual 8-second hold).</li></ul></li><li><strong>Make an Initial Grind:</strong><ul><li>Set the grinder to a starting point of <strong>15</strong>.</li><li>Adjust the grind <strong>time</strong> until the grinder dispenses exactly 18g.</li></ul></li><li><strong>Pull the First Shot:</strong><ul><li>Brew manually, stopping at <strong>36g</strong> of liquid in the cup. Note the <strong>total brew time</strong>.</li></ul></li><li><strong>Taste and Diagnose:</strong><ul><li><strong>Fast & Sour? (&lt;25s):</strong> Grind is too coarse.</li><li><strong>Slow & Bitter? (>32s):</strong> Grind is too fine.</li></ul></li><li><strong>Make ONE Adjustment - THE GRIND SIZE:</strong><ul><li>If fast/sour, adjust the grind <strong>finer</strong> (e.g., from 15 down to 13).</li><li>If slow/bitter, adjust the grind <strong>coarser</strong> (e.g., from 15 up to 17).</li></ul></li><li><strong>Re-adjust and Repeat:</strong><ul><li>After changing the grind setting, <strong>purge</strong> a small amount of coffee.</li><li>Re-weigh your next dose and <strong>adjust the grind time</strong> to get back to exactly 18g.</li><li>Pull another 36g shot. Repeat this process until your shot tastes balanced and the time falls roughly between <strong>25-32 seconds</strong>.</li></ul></li></ol><p>Happy brewing! With patience and this systematic approach, you&rsquo;ll be pulling consistently delicious espresso shots from your Breville Barista Pro in no time.</p></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/flashing-jetson-orin-nano-in-virtualized-environments/index.html
+++ b/posts/flashing-jetson-orin-nano-in-virtualized-environments/index.html
@@ -166,6 +166,6 @@ Flashing NVIDIA Jetson devices remotely presents unique challenges when the host
 <a class=heading-link href=#references><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h2><ul><li>NVIDIA Jetson Linux Developer Guide: <a href=https://docs.nvidia.com/jetson/ class=external-link target=_blank rel=noopener>https://docs.nvidia.com/jetson/</a></li><li>NVIDIA SDK Manager Documentation: <a href=https://developer.nvidia.com/sdk-manager class=external-link target=_blank rel=noopener>https://developer.nvidia.com/sdk-manager</a></li><li>Proxmox VE PCI Passthrough: <a href=https://pve.proxmox.com/wiki/PCI_Passthrough class=external-link target=_blank rel=noopener>https://pve.proxmox.com/wiki/PCI_Passthrough</a></li><li>Linux USB Networking Drivers: <a href=https://www.kernel.org/doc/html/latest/usb/ class=external-link target=_blank rel=noopener>https://www.kernel.org/doc/html/latest/usb/</a></li><li>QEMU USB Documentation: <a href=https://www.qemu.org/docs/master/system/devices/usb.html class=external-link target=_blank rel=noopener>https://www.qemu.org/docs/master/system/devices/usb.html</a></li><li>LXC Container Configuration: <a href=https://linuxcontainers.org/lxc/manpages/man5/lxc.container.conf.5.html class=external-link target=_blank rel=noopener>https://linuxcontainers.org/lxc/manpages/man5/lxc.container.conf.5.html</a></li></ul></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/how-rvq-teaches-llms-to-see-and-hear/index.html
+++ b/posts/how-rvq-teaches-llms-to-see-and-hear/index.html
@@ -16,6 +16,6 @@ The answer lies in creating a universal language—a bridge between the continuo
 <a class=heading-link href=#integrating-with-llms-two-powerful-paths-to-multi-modality><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h4><p>Once we have a contrastively-trained VQ-AE, we can use its output to give LLMs the ability to see and hear. There are two primary strategies for this.</p><p><strong>Path 1: The Tokenizer Approach - Teaching the LLM a New Language</strong></p><p>This path treats the RVQ IDs as a new vocabulary. It’s a two-stage process ideal for high-fidelity content generation.</p><ol><li><strong>Create a Neural Codec:</strong> The trained VQ-AE serves as a powerful &ldquo;codec.&rdquo; You can take any piece of media (e.g., a song) and use the codec to compress it into a sequence of discrete RVQ tokens (e.g., <code>[8, 5, 4, 1, 8, 5, 9, 2, ...]</code>).</li><li><strong>Train a Generative LLM:</strong> A new Transformer model is trained auto-regressively on a massive dataset of these media-derived tokens. Its sole purpose is to learn the patterns and predict the next token in a sequence.</li></ol><p><strong>Use Case:</strong> This is the architecture behind models like Meta&rsquo;s MusicGen. A user provides a text prompt, which conditions the Transformer to generate a new sequence of RVQ tokens. These tokens are then fed to the VQ-AE&rsquo;s decoder to synthesize the final audio waveform.</p><p><strong>Path 2: The Adapter Approach - Translating for a Language Expert</strong></p><p>This path is used to augment a powerful, pre-trained, text-only LLM without the astronomical cost of retraining it.</p><ol><li><strong>Freeze the LLM:</strong> A massive, pre-trained LLM (like LLaMA) is frozen. Its deep language understanding is preserved.</li><li><strong>Use the Pre-Quantized Embedding:</strong> Instead of using the discrete RVQ tokens, we take the rich, continuous embedding vector produced by our media encoder <em>just before</em> it enters the RVQ module.</li><li><strong>Train a Small Adapter:</strong> A small, lightweight projection layer (or &ldquo;adapter&rdquo;) is trained. Its only job is to translate the media embedding into a vector that has the same format and structure as the LLM&rsquo;s own word embeddings. It learns to map visual concepts to their corresponding &ldquo;word&rdquo; concepts in the LLM&rsquo;s latent space.</li></ol><p><strong>Use Case:</strong> This is the principle behind models like Google&rsquo;s Flamingo. To answer a question about an image, the image is passed through the media encoder and adapter. The resulting &ldquo;vision-as-a-word&rdquo; vector is inserted into the prompt sequence alongside the text tokens. The frozen LLM can now &ldquo;reason&rdquo; about the visual input because it has been translated into a format it already understands.</p></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/index.html
+++ b/posts/index.html
@@ -12,6 +12,6 @@
 <a class=title href=/posts/unifi-vlan-migration-to-zone-based-architecture/>UniFi VLAN Migration to Zone-Based Architecture</a></li><li><span class=date>August 19, 2025</span>
 <a class=title href=/posts/quantization-in-llms/>Quantization in LLMs</a></li></ul><ul class=pagination><li>1</li><li><a href=/posts/page/2/>2</a></li><li class=hidden><a href=/posts/page/2/>&#8250;</a></li><li><a href=/posts/page/2/>&#187;</a></li></ul></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/jellyfin-sso-with-authentik/index.html
+++ b/posts/jellyfin-sso-with-authentik/index.html
@@ -69,6 +69,6 @@ Do not rely on header forwarding magic. Force the scheme in the plugin configura
 <a class=heading-link href=#references><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h3><ul><li>Jellyfin SSO Plugin Repository: <code>https://github.com/9p4/jellyfin-plugin-sso</code></li><li>Authentik Documentation: <code>https://goauthentik.io/docs/providers/oauth2/</code></li><li>Jellyfin API Documentation: <code>https://api.jellyfin.org/</code></li></ul></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/mixture-of-experts-moe-models-challenges-solutions-in-practice/index.html
+++ b/posts/mixture-of-experts-moe-models-challenges-solutions-in-practice/index.html
@@ -42,6 +42,6 @@ Sparse MoE models, despite only activating a few experts per token, possess a ve
 MoE models offer significant advantages in terms of model capacity and computational efficiency, but their unique sparse activation pattern introduces challenges in training and fine-tuning. Overcoming non-differentiability in routing and ensuring balanced expert utilization are crucial for effective pre-training. During fine-tuning, managing the model&rsquo;s vast parameter count to prevent overfitting on smaller datasets requires either strategic parameter freezing or access to very large and diverse fine-tuning data.
 The <strong>Top-K routing</strong> mechanism, as illustrated in the provided image, is a core component in many modern Mixture-of-Experts (MoE) models. It involves selecting a fixed number (<code>K</code>) of experts for each input based on relevance scores.</p><hr><p><strong>Traditional Top-K (Deterministic Selection):</strong></p><ul><li><strong>How it works:</strong><ol><li>Calculate relevance scores (<code>s_{i,t}</code>) for each expert <code>i</code> and input <code>t</code>.</li><li>Identify the <code>K</code> experts with the highest scores.</li><li>Experts <em>within</em> the Top-K are assigned their scores (<code>g_{i,t} = s_{i,t}</code>).</li><li>Experts <em>outside</em> the Top-K are assigned a score of <code>0</code> (<code>g_{i,t} = 0</code>).</li><li>The output is a weighted sum of the selected experts&rsquo; outputs.</li></ol></li><li><strong>Pros:</strong> Predictable, deterministic, selects the &ldquo;best&rdquo; experts based on current scores.</li><li><strong>Cons:</strong> Can lead to expert imbalance, where a few popular experts are always chosen, starving others of training.</li></ul><p><strong>Alternative: Sampling from Softmax (Probabilistic Selection):</strong></p><ul><li><strong>How it works:</strong><ol><li>Calculate relevance scores (<code>s_{i,t}</code>) which are treated as probabilities (after softmax).</li><li><strong>Randomly sample</strong> <code>K</code> unique expert indices from the distribution defined by these probabilities.</li><li>Selected experts contribute; unselected experts do not.</li></ol></li><li><strong>Why it&rsquo;s suggested:</strong><ul><li><strong>Load Balancing:</strong> Prevents expert collapse by ensuring all experts get a chance to be selected, even those with slightly lower scores. This promotes more even training across the entire expert pool.</li><li><strong>Diversity & Exploration:</strong> Introduces randomness, potentially leading to better generalization and robustness by exploring different expert combinations.</li></ul></li><li><strong>Pros:</strong> Better load balancing, prevents expert starvation, encourages exploration.</li><li><strong>Cons:</strong> Stochastic (non-deterministic routing), can make debugging harder, might not pick the absolute &ldquo;best&rdquo; expert in a single instance (but better for long-term training).</li></ul><p><strong>Key Takeaway:</strong> While deterministic Top-K is simpler and directly picks the &ldquo;highest-scoring&rdquo; experts, sampling from the softmax offers a more robust training dynamic by ensuring that all experts receive training data, thereby preventing some experts from becoming unused (&ldquo;dead experts&rdquo;).</p><hr></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/open-webui-openai-websearch/index.html
+++ b/posts/open-webui-openai-websearch/index.html
@@ -84,6 +84,6 @@ This post documents the final setup, the hotfix script that keeps LiteLLM honest
 <a class=heading-link href=#references><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h2><ul><li><a href=https://docs.litellm.ai/docs/proxy/openai_responses class=external-link target=_blank rel=noopener>LiteLLM Documentation - OpenAI Responses API Bridge</a></li><li><a href=https://platform.openai.com/docs/api-reference/responses class=external-link target=_blank rel=noopener>OpenAI Documentation - Responses API</a></li><li><a href=https://github.com/BerriAI/litellm/issues/13042 class=external-link target=_blank rel=noopener>LiteLLM GitHub Issue #13042</a></li><li><a href=https://docs.openwebui.com/ class=external-link target=_blank rel=noopener>Open WebUI Documentation</a></li><li><a href=https://ericxliu.me/hotfix.py class=external-link target=_blank rel=noopener>The hotfix.py Script</a></li></ul></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/openwrt-mwan3-wireguard-endpoint-exclusion/index.html
+++ b/posts/openwrt-mwan3-wireguard-endpoint-exclusion/index.html
@@ -96,6 +96,6 @@ When using WireGuard together with MWAN3 on OpenWrt, the tunnel can fail to esta
 <a class=heading-link href=#references><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h3><ul><li>Session log and configs (internal): <code>~/Downloads/chat-MWAN3 WireGuard Routing Fix 🌐.txt</code></li><li>OpenWrt MWAN3 documentation: <code>https://openwrt.org/docs/guide-user/network/wan/multiwan/mwan3</code></li><li>WireGuard documentation: <code>https://www.wireguard.com/</code></li><li>OpenWrt WireGuard (UG): <code>https://openwrt.org/docs/guide-user/services/vpn/wireguard</code></li></ul></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/page/2/index.html
+++ b/posts/page/2/index.html
@@ -12,6 +12,6 @@
 <a class=title href=/posts/transformer-s-core-mechanics/>Transformer's Core Mechanics</a></li><li><span class=date>October 26, 2020</span>
 <a class=title href=/posts/useful/>Some useful files</a></li></ul><ul class=pagination><li><a href=/posts/>&#171;</a></li><li class=hidden><a href=/posts/>&#8249;</a></li><li><a href=/posts/>1</a></li><li>2</li></ul></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/ppo-for-language-models/index.html
+++ b/posts/ppo-for-language-models/index.html
@@ -23,6 +23,6 @@ where <code>δ_t = r_t + γV(s_{t+1}) - V(s_t)</code></p><ul><li><strong>γ (gam
 <a class=heading-link href=#the-full-training-loop><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h3><p>Now, we can assemble the entire process into a clear, iterative loop:</p><ol><li><strong>Collect</strong>: The current Actor policy <code>π_k</code> generates responses to a batch of prompts. These experiences—<code>(state, action, probability, reward, value)</code>—are stored in an <strong>Experience Buffer</strong>.</li><li><strong>Calculate</strong>: Once the buffer is full, we use the collected data to compute the advantage estimates <code>Â_t</code> for every single token-generation step.</li><li><strong>Optimize</strong>: For a few epochs, we repeatedly sample mini-batches from the buffer and update the Actor and Critic models. The Actor is updated using the combined <code>PPO-clip Loss</code> and <code>LM Loss</code>. The Critic is updated to improve its value predictions.</li><li><strong>Flush and Repeat</strong>: After the optimization phase, the entire experience buffer is discarded. The data is now &ldquo;stale&rdquo; because our policy has changed. The newly updated policy <code>π_{k+1}</code> becomes the new Actor, and we return to step 1 to collect fresh data.</li></ol><p>This cycle of collection and optimization allows the language model to gradually and safely steer its behavior towards human-defined goals, creating the helpful and aligned AI assistants we interact with today.</p><hr><p><strong>References:</strong></p><ol><li>Schulman, J., Wolski, F., Dhariwal, P., Radford, A., & Klimov, O. (2017). <em>Proximal Policy Optimization Algorithms</em>. arXiv preprint arXiv:1707.06347.</li><li>Schulman, J., Moritz, P., Levine, S., Jordan, M., & Abbeel, P. (2015). <em>High-Dimensional Continuous Control Using Generalized Advantage Estimation</em>. arXiv preprint arXiv:1506.02438.</li><li>Ouyang, L., et al. (2022). <em>Training language models to follow instructions with human feedback</em>. Advances in Neural Information Processing Systems 35.</li></ol></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/quantization-in-llms/index.html
+++ b/posts/quantization-in-llms/index.html
--- a/posts/secure-boot-dkms-and-mok-on-proxmox-debian/index.html
+++ b/posts/secure-boot-dkms-and-mok-on-proxmox-debian/index.html
@@ -57,6 +57,6 @@ nvidia-smi failed to communicate with the NVIDIA driver modprobe nvidia → “K
 <a class=heading-link href=#references><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h3><ul><li>Proxmox Secure Boot setup (shim + MOK, EFI vars, DKMS): <a href=https://pve.proxmox.com/wiki/Secure_Boot_Setup#Setup_instructions_for_shim_+_MOK_variant class=external-link target=_blank rel=noopener>Proxmox docs</a></li></ul></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/supabase-deep-dive/index.html
+++ b/posts/supabase-deep-dive/index.html
@@ -88,6 +88,6 @@ Supabase enters this space with a radically different philosophy: transparency.
 <a class=heading-link href=#final-words-of-advice><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h3><ul><li><strong>Frontend Freedom:</strong> Supabase is frontend-agnostic, but meta-frameworks like <strong>Next.js</strong> and <strong>SvelteKit</strong> offer a &ldquo;golden path&rdquo; with Auth Helpers that simplify server-side rendering and data fetching.</li><li><strong>Embrace the CLI:</strong> Use the Supabase CLI for a professional, safe, and repeatable development workflow. Don&rsquo;t manage your production schema by clicking in the UI.</li><li><strong>Know Your Keys:</strong> Use the public <code>anon</code> key in the browser. Guard the secret <code>service_role</code> key and only use it in secure server environments like Edge Functions.</li><li><strong>Indexes Matter:</strong> For fast queries on large tables, <code>CREATE INDEX</code> on frequently queried columns. Performance is not automatic.</li></ul><p>By understanding these principles, you can leverage Supabase not as a simple BaaS, but as a powerful, transparent, and scalable platform for building next-generation applications on the solid foundation of PostgreSQL.</p></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/t5-the-transformer-that-zigged-when-others-zagged-an-architectural-deep-dive/index.html
+++ b/posts/t5-the-transformer-that-zigged-when-others-zagged-an-architectural-deep-dive/index.html
@@ -28,6 +28,6 @@ But to truly understand the field, we must look at the pivotal models that explo
 <a class=heading-link href=#conclusion-the-lasting-legacy><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h3><p>T5 represents a different evolutionary branch in the Transformer family tree. While the field has largely converged on the decoder-only architecture for its scalability in general-purpose models, T5&rsquo;s design remains a masterclass in purpose-built engineering.</p><p>Its text-to-text framework was revolutionary, its encoder-decoder structure is still a go-to for tasks like translation, and its refined T5 v1.1 architecture laid the groundwork for many of the stability and efficiency tricks we see in today&rsquo;s state-of-the-art models. T5 is more than just a model; it&rsquo;s a crucial case study in the architectural trade-offs that continue to shape the future of artificial intelligence.</p></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/technical-deep-dive-llm-categorization/index.html
+++ b/posts/technical-deep-dive-llm-categorization/index.html
@@ -71,6 +71,6 @@ It turned out to be a syntax error in my arguments passed to the <code>Trainer</
 <a class=heading-link href=#conclusion><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h2><p>We often think we need 70B parameter models for everything. usage shows that for a specific, well-defined task with consistent formatting, a <strong>270M parameter model</strong>—fine-tuned on high-quality, distilled data—can punch way above its weight class.</p><p>The key was <strong>data quality over quantity</strong>. By using the commercial model to &ldquo;verify&rdquo; my historical data, I created a dataset that was cleaner than either source alone.</p></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/the-convergence-of-fast-weights-linear-attention-and-state-space-models/index.html
+++ b/posts/the-convergence-of-fast-weights-linear-attention-and-state-space-models/index.html
@@ -24,6 +24,6 @@ This article explores the mathematical equivalence between Hinton’s concept of
 <a class=heading-link href=#references><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h3><ol><li><strong>Hinton, G. E., & Plaut, D. C. (1987).</strong> &ldquo;Using Fast Weights to Deblur Old Memories.&rdquo; <em>Proceedings of the 9th Annual Conference of the Cognitive Science Society.</em></li><li><strong>Ba, J., Hinton, G. E., et al. (2016).</strong> &ldquo;Using Fast Weights to Attend to the Recent Past.&rdquo; <em>Advances in Neural Information Processing Systems (NeurIPS).</em></li><li><strong>Katharopoulos, A., et al. (2020).</strong> &ldquo;Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention.&rdquo; <em>International Conference on Machine Learning (ICML).</em></li><li><strong>Gu, A., & Dao, T. (2023).</strong> &ldquo;Mamba: Linear-Time Sequence Modeling with Selective State Spaces.&rdquo; <em>arXiv preprint arXiv:2312.00752.</em></li><li><strong>Vaswani, A., et al. (2017).</strong> &ldquo;Attention Is All You Need.&rdquo; <em>Advances in Neural Information Processing Systems (NeurIPS).</em></li></ol></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/transformer-s-core-mechanics/index.html
+++ b/posts/transformer-s-core-mechanics/index.html
@@ -34,6 +34,6 @@ In deep learning, a &ldquo;channel&rdquo; can be thought of as a feature dimensi
 <a class=heading-link href=#references><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h3><ol><li>Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., &mldr; & Polosukhin, I. (2017). Attention is all you need. <em>Advances in neural information processing systems</em>, 30.</li><li>Shazeer, N., Mirhoseini, A., Maziarz, K., Davis, A., Le, Q., Hinton, G., & Dean, J. (2017). Outrageously large neural networks: The sparsely-gated mixture-of-experts layer. <em>arXiv preprint arXiv:1701.06538</em>.</li><li>Ainslie, J., Ontanon, J., Cakka, E., Dosovitskiy, A., & Le, Q. V. (2023). GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints. <em>arXiv preprint arXiv:2305.13245</em>.</li></ol></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/unifi-vlan-migration-to-zone-based-architecture/index.html
+++ b/posts/unifi-vlan-migration-to-zone-based-architecture/index.html
@@ -26,6 +26,6 @@ This article documents that journey. It details the pitfalls encountered, the co
 <a class=heading-link href=#references><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h3><ul><li><a href=https://help.ui.com/hc/en-us/articles/7258465146519-Troubleshooting-UniFi-Device-Connectivity class=external-link target=_blank rel=noopener>Troubleshooting UniFi Device Connectivity</a></li><li><a href=https://help.ui.com/hc/en-us/articles/9592924981911-Virtual-Network-VLAN-Troubleshooting class=external-link target=_blank rel=noopener>Virtual Network (VLAN) Troubleshooting</a></li></ul></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/useful/index.html
+++ b/posts/useful/index.html
@@ -7,6 +7,6 @@ rootCA.pem
 </time></span><span class=reading-time><i class="fa-solid fa-clock" aria-hidden=true></i>
 One-minute read</span></div></div></header><div class=post-content><ul><li><a href=/rootCA.crt>rootCA.pem</a></li></ul></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>
--- a/posts/vattention/index.html
+++ b/posts/vattention/index.html
@@ -29,6 +29,6 @@ The GPU TLB hierarchy is sensitive to page sizes.</p><ul><li><strong>4KB Pages:<
 <a class=heading-link href=#references><i class="fa-solid fa-link" aria-hidden=true title="Link to heading"></i>
 <span class=sr-only>Link to heading</span></a></h3><p>[1] R. Prabhu et al., &ldquo;vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention,&rdquo; in <em>Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS &lsquo;25)</em>, 2025.</p></div><footer><div id=disqus_thread></div><script>window.disqus_config=function(){},function(){if(["localhost","127.0.0.1"].indexOf(window.location.hostname)!=-1){document.getElementById("disqus_thread").innerHTML="Disqus comments not available by default when the website is previewed locally.";return}var t=document,e=t.createElement("script");e.async=!0,e.src="//ericxliu-me.disqus.com/embed.js",e.setAttribute("data-timestamp",+new Date),(t.head||t.body).appendChild(e)}(),document.addEventListener("themeChanged",function(){document.readyState=="complete"&&DISQUS.reset({reload:!0,config:disqus_config})})</script></footer></article><link rel=stylesheet href=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.css integrity=sha384-vKruj+a13U8yHIkAyGgK1J3ArTLzrFGBbBc0tDp4ad/EyewESeXE/Iv67Aj8gKZ0 crossorigin=anonymous><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/katex.min.js integrity=sha384-PwRUT/YqbnEjkZO0zZxNqcxACrXe+j766U2amXcgMg5457rve2Y7I6ZJSm2A0mS4 crossorigin=anonymous></script><script defer src=https://cdn.jsdelivr.net/npm/katex@0.16.4/dist/contrib/auto-render.min.js integrity=sha384-+VBxd3r6XgURycqtZ117nYw44OOcIax56Z4dCRWbxyPt0Koah1uHoK0o4+/RRE05 crossorigin=anonymous onload='renderMathInElement(document.body,{delimiters:[{left:"$$",right:"$$",display:!0},{left:"$",right:"$",display:!1},{left:"\\(",right:"\\)",display:!1},{left:"\\[",right:"\\]",display:!0}]})'></script></section></div><footer class=footer><section class=container>©
 2016 -
-2025
+2026
 Eric X. Liu
 <a href="https://git.ericxliu.me/eric/ericxliu-me/commit/f1178d3">[f1178d3]</a></section></footer></main><script src=/js/coder.min.6ae284be93d2d19dad1f02b0039508d9aab3180a12a06dcc71b0b0ef7825a317.js integrity="sha256-auKEvpPS0Z2tHwKwA5UI2aqzGAoSoG3McbCw73gloxc="></script><script defer src=https://static.cloudflareinsights.com/beacon.min.js data-cf-beacon='{"token": "987638e636ce4dbb932d038af74c17d1"}'></script></body></html>