39 citations · 47 across the 5 of their papers we have counts for
6 papers
MCF: A Memory- and Communication-Aware Co-Designed Framework for Long-Context LLM Inference
Huaizhi Zhang, Jiacheng Zhu, Xuqi Zhu +3
This paper presents Memory- and Communication-Aware Framework for Long-Context LLM Inference (MCF) that optimizes long-context inference through communication-aware and stage-adapt…
Mitigating scalability challenges in LUT-based neural networks via pruning optimisations
Xuqi Zhu, Huaizhi Zhang, JunKyu Lee +5
Modern deep neural networks heavily rely on a large number of multiply-accumulate operations, which constitute the predominant computational cost. To address this, Look-Up Table (L…
Vortex Feature Positioning: Bridging Tabular IIoT Data and Image-Based Deep Learning
Jong-Ik Park, Sihoon Seong, JunKyu Lee +1
Tabular data from IIoT devices are typically analyzed using decision tree-based machine learning techniques, which struggle with high-dimensional and numeric data. To overcome thes…
ROMA: Run-Time Object Detection To Maximize Real-Time Accuracy
JunKyu Lee, Blesson Varghese, Hans Vandierendonck
This paper analyzes the effects of dynamically varying video contents and detection latency on the real-time detection accuracy of a detector and proposes a new run-time accuracy v…
Resource-Efficient Deep Learning: A Survey on Model-, Arithmetic-, and Implementation-Level Techniques
JunKyu Lee, Lev Mukhanov, Amir Sabbagh Molahosseini +6
Deep learning is pervasive in our daily life, including self-driving cars, virtual assistants, social network services, healthcare services, face recognition, etc. However, deep ne…
TOD: Transprecise Object Detection to Maximise Real-Time Accuracy on the Edge
JunKyu Lee, Blesson Varghese, Roger Woods +1
Real-time video analytics on the edge is challenging as the computationally constrained resources typically cannot analyse video streams at full fidelity and frame rate, which resu…