English(EN) SPARTA: A Sparse Ternary LLM Accelerator with Adaptive Attention and Unified Datapath for Long-Context Edge Inference

SPARTA加速器通过稀疏三元设计提升LLM边缘推理性能 · 跟踪到1个来源

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-27 04:00

研究人员开发了SPARTA，这是一种新颖的加速器，专为边缘设备上稀疏三元大语言模型（LLM）的高效长上下文推理而设计。SPARTA采用面向阶段的稀疏数据流和统一的计算结构，将三元运算和注意力机制映射到共享处理单元上，从而利用各种形式的稀疏性。一个自适应的两级注意力引擎通过将KV缓存流量减少高达4.8倍，同时损失极小的质量。基准测试表明，SPARTA在生成吞吐量和能效方面显著优于NVIDIA Jetson AGX Orin和AMD Ryzen 7 9700X等现有硬件。 AI

影响使资源受限的边缘设备上能够实现更强大的LLM功能，可能扩展AI在移动和嵌入式系统中的应用。

排序理由该集群包含一篇详细介绍LLM推理新硬件架构的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 14 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍LLM推理新硬件架构的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: infra, paper; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Yanzhen Zhou, Zhengyu Han, Qunyou Liu, Zheyu Yan, Weihong Xu, Cheng Zhuo · 2026-10-27 04:00

SPARTA：一种稀疏三元大模型加速器，具有自适应注意力和统一数据通路，用于长上下文边缘推理

arXiv:2610.34824v1 Announce Type: cross Abstract: LLMs on edge devices are constrained by limited memory and bandwidth. Although 1.58-bit ternary weights greatly reduce model storage, long-context inference is increasingly bottlenecked by KV-cache traffic and attention. Existing …

本文如何被排名

报道来源 [1]

SPARTA：一种稀疏三元大模型加速器，具有自适应注意力和统一数据通路，用于长上下文边缘推理

相关实体

相关话题