29 lines
21 KiB
HTML
29 lines
21 KiB
HTML
<!doctype html><html lang=en dir=auto data-theme=auto><head><meta charset=utf-8><meta http-equiv=X-UA-Compatible content="IE=edge"><meta name=viewport content="width=device-width,initial-scale=1,shrink-to-fit=no"><meta name=robots content="index, follow"><title>SEC 2025 Digest | Publish Assistant</title><meta name=keywords content><meta name=description content="12 papers selected.
|
|
|
|
lm-Meter: Unveiling Runtime Inference Latency for On-Device Language Models
|
|
Haoxin Wang 0003
|
|
TL;DR — Provides the first detailed runtime profiling framework for on-device LLM inference, revealing key latency bottlenecks across diverse edge hardware configurations.
|
|
|
|
SLED: A Speculative LLM Decoding Framework for Efficient Edge Serving
|
|
Xiangchen Li
|
|
TL;DR — Adapts speculative decoding to edge serving constraints, reducing LLM token generation latency while respecting the tight memory and compute budgets of edge nodes."><meta name=author content="Publish Assistant"><link rel=canonical href=https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/sec-2025/><link crossorigin=anonymous href=/vincent/publish-assistant/assets/css/stylesheet.d72f07832e13c592b3edba91680bfe70f01daac396179bcace0ac36e8e0494c6.css integrity="sha256-1y8Hgy4TxZKz7bqRaAv+cPAdqsOWF5vKzgrDbo4ElMY=" rel="preload stylesheet" as=style><link rel=icon href=https://pub.sqrt.fr/vincent/publish-assistant/favicon.ico><link rel=icon type=image/png sizes=16x16 href=https://pub.sqrt.fr/vincent/publish-assistant/favicon-16x16.png><link rel=icon type=image/png sizes=32x32 href=https://pub.sqrt.fr/vincent/publish-assistant/favicon-32x32.png><link rel=apple-touch-icon href=https://pub.sqrt.fr/vincent/publish-assistant/apple-touch-icon.png><link rel=mask-icon href=https://pub.sqrt.fr/vincent/publish-assistant/safari-pinned-tab.svg><meta name=theme-color content="#2e2e33"><meta name=msapplication-TileColor content="#2e2e33"><link rel=alternate hreflang=en href=https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/sec-2025/><noscript><style>#theme-toggle,.top-link{display:none}</style><style>@media(prefers-color-scheme:dark){:root{--theme:rgb(29, 30, 32);--entry:rgb(46, 46, 51);--primary:rgb(218, 218, 219);--secondary:rgb(155, 156, 157);--tertiary:rgb(65, 66, 68);--content:rgb(196, 196, 197);--code-block-bg:rgb(46, 46, 51);--code-bg:rgb(55, 56, 62);--border:rgb(51, 51, 51);color-scheme:dark}.list{background:var(--theme)}.toc{background:var(--entry)}}</style></noscript><script>localStorage.getItem("pref-theme")==="dark"?document.querySelector("html").dataset.theme="dark":localStorage.getItem("pref-theme")==="light"?document.querySelector("html").dataset.theme="light":window.matchMedia("(prefers-color-scheme: dark)").matches?document.querySelector("html").dataset.theme="dark":document.querySelector("html").dataset.theme="light"</script><meta property="og:url" content="https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/sec-2025/"><meta property="og:site_name" content="Publish Assistant"><meta property="og:title" content="SEC 2025 Digest"><meta property="og:description" content="12 papers selected.
|
|
lm-Meter: Unveiling Runtime Inference Latency for On-Device Language Models Haoxin Wang 0003
|
|
TL;DR — Provides the first detailed runtime profiling framework for on-device LLM inference, revealing key latency bottlenecks across diverse edge hardware configurations.
|
|
SLED: A Speculative LLM Decoding Framework for Efficient Edge Serving Xiangchen Li
|
|
TL;DR — Adapts speculative decoding to edge serving constraints, reducing LLM token generation latency while respecting the tight memory and compute budgets of edge nodes."><meta property="og:locale" content="en_us"><meta property="og:type" content="article"><meta property="article:section" content="cloud-edge"><meta property="article:published_time" content="2025-01-01T00:00:00+00:00"><meta property="article:modified_time" content="2025-01-01T00:00:00+00:00"><meta name=twitter:card content="summary"><meta name=twitter:title content="SEC 2025 Digest"><meta name=twitter:description content="12 papers selected.
|
|
lm-Meter: Unveiling Runtime Inference Latency for On-Device Language Models Haoxin Wang 0003
|
|
TL;DR — Provides the first detailed runtime profiling framework for on-device LLM inference, revealing key latency bottlenecks across diverse edge hardware configurations.
|
|
SLED: A Speculative LLM Decoding Framework for Efficient Edge Serving Xiangchen Li
|
|
TL;DR — Adapts speculative decoding to edge serving constraints, reducing LLM token generation latency while respecting the tight memory and compute budgets of edge nodes."><script type=application/ld+json>{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Edge and Cloud Systems","item":"https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/"},{"@type":"ListItem","position":2,"name":"Digests","item":"https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/"},{"@type":"ListItem","position":3,"name":"SEC 2025 Digest","item":"https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/sec-2025/"}]}</script><script type=application/ld+json>{"@context":"https://schema.org","@type":"BlogPosting","headline":"SEC 2025 Digest","name":"SEC 2025 Digest","description":"12 papers selected.\nlm-Meter: Unveiling Runtime Inference Latency for On-Device Language Models Haoxin Wang 0003\nTL;DR — Provides the first detailed runtime profiling framework for on-device LLM inference, revealing key latency bottlenecks across diverse edge hardware configurations.\nSLED: A Speculative LLM Decoding Framework for Efficient Edge Serving Xiangchen Li\nTL;DR — Adapts speculative decoding to edge serving constraints, reducing LLM token generation latency while respecting the tight memory and compute budgets of edge nodes.\n","keywords":[],"articleBody":"12 papers selected.\nlm-Meter: Unveiling Runtime Inference Latency for On-Device Language Models Haoxin Wang 0003\nTL;DR — Provides the first detailed runtime profiling framework for on-device LLM inference, revealing key latency bottlenecks across diverse edge hardware configurations.\nSLED: A Speculative LLM Decoding Framework for Efficient Edge Serving Xiangchen Li\nTL;DR — Adapts speculative decoding to edge serving constraints, reducing LLM token generation latency while respecting the tight memory and compute budgets of edge nodes.\nSledgeScale: Load-Aware Dispatch and Deadline-Driven Scheduling for Scalable, Dense Serverless Computing in Edge Data Centers Xiaosu Lyu\nTL;DR — Introduces a load-aware dispatch and deadline-driven scheduler for dense serverless edge data centers, demonstrating substantial SLA compliance improvements over baseline policies.\nWarping the Edge: Enabling Instant Mobility for Stateful Applications over 5G and Beyond Mukhtiar Ahmad\nTL;DR — Achieves near-instantaneous stateful application migration across 5G edge nodes by combining memory snapshotting with network-layer forwarding continuity.\nUncertainty-Aware RL-Based Scheduling of Multi-DNN Workloads on Edge MPSoCs Soroush Heidari\nTL;DR — Uses uncertainty-aware reinforcement learning to schedule concurrent DNN workloads on heterogeneous edge MPSoCs, reducing deadline misses under dynamic arrival patterns.\nSEEB-GPU: Early-Exit Aware Scheduling and Batching for Edge GPU Inference Srinivasan Subramaniyan\nTL;DR — Exploits early-exit branching in DNN inference to build an adaptive batching and scheduling policy for edge GPUs that cuts average latency without sacrificing throughput.\nElastoformer: Enabling Dynamic Adaptivity via Elastic Model Transformation Sudaksh Kalra\nTL;DR — Proposes elastic transformer transformations that resize model capacity at runtime to match available edge resources, enabling continuous inference under fluctuating conditions.\nPlatformX: An End-to-End Transferable Platform for Energy-Efficient Neural Architecture Search Xiaolong Tu\nTL;DR — Presents a transferable NAS platform that searches for energy-efficient DNN architectures deployable across heterogeneous edge targets with minimal re-search overhead.\nBayes-Split-Edge: Bayesian Optimization for Constrained Collaborative Inference in Wireless Edge Systems Fatemeh Zahra Safaeipour\nTL;DR — Applies Bayesian optimization to find optimal split points for collaborative inference in wireless edge systems, accounting for dynamic channel and computation constraints.\nEnergy-efficient DNN Dividing Technique for Latency Optimization in Dynamic Mobile Edge Networks Eldiyar Zhantileuov\nTL;DR — Develops a DNN partitioning strategy for mobile edge networks that minimizes end-to-end latency while satisfying energy budgets under time-varying link conditions.\nLLM-Driven Auto Configuration for Transient IoT Device Collaboration Hetvi Shastri\nTL;DR — Leverages LLMs to automate the configuration of transient IoT device coalitions, reducing manual setup overhead and adapting collaboration policies to changing device membership.\nfReeLoaders: An IoT Ecosystem for Real-Time Deadline-Driven Task Scheduling using Reinforcement Learning Marshall Clyburn\nTL;DR — Builds a reinforcement-learning scheduler for IoT ecosystems that meets real-time task deadlines by exploiting opportunistic idle capacity across heterogeneous edge devices.\n","wordCount":"433","inLanguage":"en","datePublished":"2025-01-01T00:00:00Z","dateModified":"2025-01-01T00:00:00Z","author":{"@type":"Person","name":"Publish Assistant"},"mainEntityOfPage":{"@type":"WebPage","@id":"https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/sec-2025/"},"publisher":{"@type":"Organization","name":"Publish Assistant","logo":{"@type":"ImageObject","url":"https://pub.sqrt.fr/vincent/publish-assistant/favicon.ico"}}}</script></head><body id=top><header class=header><nav class=header-nav><div class=logo><a href=https://pub.sqrt.fr/vincent/publish-assistant/ accesskey=h title="Publish Assistant (Alt + H)">Publish Assistant</a>
|
|
<span class=logo-sep>/</span>
|
|
<a class=logo-topic href=/vincent/publish-assistant/cloud-edge/ title="Edge and Cloud Systems">Edge and Cloud Systems</a><div class=logo-switches><button id=theme-toggle class=theme-toggle accesskey=t title="(Alt + T)" aria-label="Toggle theme">
|
|
<svg class="moon" width="18" height="18" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round"><path d="M21 12.79A9 9 0 1111.21 3 7 7 0 0021 12.79z"/></svg>
|
|
<svg class="sun" width="18" height="18" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round"><circle cx="12" cy="12" r="5"/><line x1="12" y1="1" x2="12" y2="3"/><line x1="12" y1="21" x2="12" y2="23"/><line x1="4.22" y1="4.22" x2="5.64" y2="5.64"/><line x1="18.36" y1="18.36" x2="19.78" y2="19.78"/><line x1="1" y1="12" x2="3" y2="12"/><line x1="21" y1="12" x2="23" y2="12"/><line x1="4.22" y1="19.78" x2="5.64" y2="18.36"/><line x1="18.36" y1="5.64" x2="19.78" y2="4.22"/></svg></button></div></div><ul id=menu class=menu><li><a href=/vincent/publish-assistant/cloud-edge/venues/ title=Venues><span>Venues</span></a></li><li><a href=/vincent/publish-assistant/cloud-edge/calendar/ title=Calendar><span>Calendar</span></a></li><li><a href=/vincent/publish-assistant/cloud-edge/digests/ title=Digests><span class=active>Digests</span></a></li></ul></nav></header><main class=main><article class=post-single><header class=post-header><nav class=breadcrumbs role=navigation aria-label=Breadcrumb><a href=/vincent/publish-assistant/cloud-edge/digests/>Digests</a>
|
|
<svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="feather feather-chevron-right"><polyline points="9 18 15 12 9 6"/></svg></nav><h1 class="post-title entry-hint-parent">SEC 2025 Digest</h1><div class=post-meta><span title='2025-01-01 00:00:00 +0000 UTC'>January 1, 2025</span> · <span>Publish Assistant</span></div></header><div class="post-content md-content"><p>12 papers selected.</p><hr><h3 id=lm-meter-unveiling-runtime-inference-latency-for-on-device-language-models>lm-Meter: Unveiling Runtime Inference Latency for On-Device Language Models<a hidden class=anchor aria-hidden=true href=#lm-meter-unveiling-runtime-inference-latency-for-on-device-language-models>#</a></h3><p><em>Haoxin Wang 0003</em></p><p><strong>TL;DR</strong> — Provides the first detailed runtime profiling framework for on-device LLM inference, revealing key latency bottlenecks across diverse edge hardware configurations.</p><hr><h3 id=sled-a-speculative-llm-decoding-framework-for-efficient-edge-serving>SLED: A Speculative LLM Decoding Framework for Efficient Edge Serving<a hidden class=anchor aria-hidden=true href=#sled-a-speculative-llm-decoding-framework-for-efficient-edge-serving>#</a></h3><p><em>Xiangchen Li</em></p><p><strong>TL;DR</strong> — Adapts speculative decoding to edge serving constraints, reducing LLM token generation latency while respecting the tight memory and compute budgets of edge nodes.</p><hr><h3 id=sledgescale-load-aware-dispatch-and-deadline-driven-scheduling-for-scalable-dense-serverless-computing-in-edge-data-centers>SledgeScale: Load-Aware Dispatch and Deadline-Driven Scheduling for Scalable, Dense Serverless Computing in Edge Data Centers<a hidden class=anchor aria-hidden=true href=#sledgescale-load-aware-dispatch-and-deadline-driven-scheduling-for-scalable-dense-serverless-computing-in-edge-data-centers>#</a></h3><p><em>Xiaosu Lyu</em></p><p><strong>TL;DR</strong> — Introduces a load-aware dispatch and deadline-driven scheduler for dense serverless edge data centers, demonstrating substantial SLA compliance improvements over baseline policies.</p><hr><h3 id=warping-the-edge-enabling-instant-mobility-for-stateful-applications-over-5g-and-beyond>Warping the Edge: Enabling Instant Mobility for Stateful Applications over 5G and Beyond<a hidden class=anchor aria-hidden=true href=#warping-the-edge-enabling-instant-mobility-for-stateful-applications-over-5g-and-beyond>#</a></h3><p><em>Mukhtiar Ahmad</em></p><p><strong>TL;DR</strong> — Achieves near-instantaneous stateful application migration across 5G edge nodes by combining memory snapshotting with network-layer forwarding continuity.</p><hr><h3 id=uncertainty-aware-rl-based-scheduling-of-multi-dnn-workloads-on-edge-mpsocs>Uncertainty-Aware RL-Based Scheduling of Multi-DNN Workloads on Edge MPSoCs<a hidden class=anchor aria-hidden=true href=#uncertainty-aware-rl-based-scheduling-of-multi-dnn-workloads-on-edge-mpsocs>#</a></h3><p><em>Soroush Heidari</em></p><p><strong>TL;DR</strong> — Uses uncertainty-aware reinforcement learning to schedule concurrent DNN workloads on heterogeneous edge MPSoCs, reducing deadline misses under dynamic arrival patterns.</p><hr><h3 id=seeb-gpu-early-exit-aware-scheduling-and-batching-for-edge-gpu-inference>SEEB-GPU: Early-Exit Aware Scheduling and Batching for Edge GPU Inference<a hidden class=anchor aria-hidden=true href=#seeb-gpu-early-exit-aware-scheduling-and-batching-for-edge-gpu-inference>#</a></h3><p><em>Srinivasan Subramaniyan</em></p><p><strong>TL;DR</strong> — Exploits early-exit branching in DNN inference to build an adaptive batching and scheduling policy for edge GPUs that cuts average latency without sacrificing throughput.</p><hr><h3 id=elastoformer-enabling-dynamic-adaptivity-via-elastic-model-transformation>Elastoformer: Enabling Dynamic Adaptivity via Elastic Model Transformation<a hidden class=anchor aria-hidden=true href=#elastoformer-enabling-dynamic-adaptivity-via-elastic-model-transformation>#</a></h3><p><em>Sudaksh Kalra</em></p><p><strong>TL;DR</strong> — Proposes elastic transformer transformations that resize model capacity at runtime to match available edge resources, enabling continuous inference under fluctuating conditions.</p><hr><h3 id=platformx-an-end-to-end-transferable-platform-for-energy-efficient-neural-architecture-search>PlatformX: An End-to-End Transferable Platform for Energy-Efficient Neural Architecture Search<a hidden class=anchor aria-hidden=true href=#platformx-an-end-to-end-transferable-platform-for-energy-efficient-neural-architecture-search>#</a></h3><p><em>Xiaolong Tu</em></p><p><strong>TL;DR</strong> — Presents a transferable NAS platform that searches for energy-efficient DNN architectures deployable across heterogeneous edge targets with minimal re-search overhead.</p><hr><h3 id=bayes-split-edge-bayesian-optimization-for-constrained-collaborative-inference-in-wireless-edge-systems>Bayes-Split-Edge: Bayesian Optimization for Constrained Collaborative Inference in Wireless Edge Systems<a hidden class=anchor aria-hidden=true href=#bayes-split-edge-bayesian-optimization-for-constrained-collaborative-inference-in-wireless-edge-systems>#</a></h3><p><em>Fatemeh Zahra Safaeipour</em></p><p><strong>TL;DR</strong> — Applies Bayesian optimization to find optimal split points for collaborative inference in wireless edge systems, accounting for dynamic channel and computation constraints.</p><hr><h3 id=energy-efficient-dnn-dividing-technique-for-latency-optimization-in-dynamic-mobile-edge-networks>Energy-efficient DNN Dividing Technique for Latency Optimization in Dynamic Mobile Edge Networks<a hidden class=anchor aria-hidden=true href=#energy-efficient-dnn-dividing-technique-for-latency-optimization-in-dynamic-mobile-edge-networks>#</a></h3><p><em>Eldiyar Zhantileuov</em></p><p><strong>TL;DR</strong> — Develops a DNN partitioning strategy for mobile edge networks that minimizes end-to-end latency while satisfying energy budgets under time-varying link conditions.</p><hr><h3 id=llm-driven-auto-configuration-for-transient-iot-device-collaboration>LLM-Driven Auto Configuration for Transient IoT Device Collaboration<a hidden class=anchor aria-hidden=true href=#llm-driven-auto-configuration-for-transient-iot-device-collaboration>#</a></h3><p><em>Hetvi Shastri</em></p><p><strong>TL;DR</strong> — Leverages LLMs to automate the configuration of transient IoT device coalitions, reducing manual setup overhead and adapting collaboration policies to changing device membership.</p><hr><h3 id=freeloaders-an-iot-ecosystem-for-real-time-deadline-driven-task-scheduling-using-reinforcement-learning>fReeLoaders: An IoT Ecosystem for Real-Time Deadline-Driven Task Scheduling using Reinforcement Learning<a hidden class=anchor aria-hidden=true href=#freeloaders-an-iot-ecosystem-for-real-time-deadline-driven-task-scheduling-using-reinforcement-learning>#</a></h3><p><em>Marshall Clyburn</em></p><p><strong>TL;DR</strong> — Builds a reinforcement-learning scheduler for IoT ecosystems that meets real-time task deadlines by exploiting opportunistic idle capacity across heterogeneous edge devices.</p></div><footer class=post-footer><ul class=post-tags></ul><nav class=paginav><a class=prev href=https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/sc-2025/><span class=title>« Prev</span>
|
|
<span>SC 2025 Digest</span>
|
|
</a><a class=next href=https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/socc-2025/><span class=title>Next »</span>
|
|
<span>SoCC 2025 Digest</span></a></nav></footer></article></main><footer class=footer><span>© 2026 <a href=https://pub.sqrt.fr/vincent/publish-assistant/>Publish Assistant</a></span> ·
|
|
<span>Powered by
|
|
<a href="https://gohugo.io/?utm_source=papermod" rel=noopener target=_blank>Hugo</a> &
|
|
<a href=https://github.com/adityatelange/hugo-PaperMod/ rel=noopener target=_blank>PaperMod</a></span></footer><a href=#top id=top-link class="top-link hidden" aria-label="go to top" title="Go to Top (Alt + G)" accesskey=g><svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="feather feather-chevrons-up"><polyline points="17 11 12 6 7 11"/><polyline points="17 18 12 13 7 18"/></svg>
|
|
</a><script>let menu=document.getElementById("menu");if(menu){const e=localStorage.getItem("menu-scroll-position");e&&(menu.scrollLeft=parseInt(e,10)),menu.onscroll=function(){localStorage.setItem("menu-scroll-position",menu.scrollLeft)}}document.querySelectorAll('a[href^="#"]').forEach(e=>{e.addEventListener("click",function(e){e.preventDefault();var t=this.getAttribute("href").substr(1);window.matchMedia("(prefers-reduced-motion: reduce)").matches?document.querySelector(`[id='${decodeURIComponent(t)}']`).scrollIntoView():document.querySelector(`[id='${decodeURIComponent(t)}']`).scrollIntoView({behavior:"smooth"}),t==="top"?history.replaceState(null,null," "):history.pushState(null,null,`#${t}`)})})</script><script>var toplink=document.getElementById("top-link");window.onscroll=function(){const e=window.innerHeight;document.body.scrollTop>e||document.documentElement.scrollTop>e?toplink.classList.remove("hidden"):toplink.classList.add("hidden")}</script><script>document.getElementById("theme-toggle").addEventListener("click",()=>{const e=document.querySelector("html");e.dataset.theme==="dark"?(e.dataset.theme="light",localStorage.setItem("pref-theme","light")):(e.dataset.theme="dark",localStorage.setItem("pref-theme","dark"))})</script></body></html> |