person · community record
Woosuk Kwon
A connected record in the developer community graph.
Checking node ownership…
01
Connections
20 relationships
appeared_onA User-Centric Perspective on LLM Inference | AM Podcast #3talk ↗appeared_onInferact: Building the Infrastructure That Runs Modern AItalk ↗authoredA Fast Post-Training Pruning Framework for Transformerspaper ↗authoredAPEX: An Extensible and Dynamism-Aware Simulator for Automated Parallel Execution in LLM Servingpaper ↗authoredEfficient Memory Management for Large Language Model Serving with PagedAttentionpaper ↗authoredGraphene: Strong yet Lightweight Row Hammer Protectionpaper ↗authoredJenga: Effective Memory Management for Serving LLM with Heterogeneitypaper ↗authoredLearned Token Pruning for Transformerspaper ↗authoredNimble: Lightweight and Parallel GPU Task Scheduling for Deep Learningpaper ↗authoredSkyPilot: An Intercloud Broker for Sky Computingpaper ↗authoredTurboSpec: Closed-loop Speculation Control System for Optimizing LLM Serving Goodputpaper ↗inventor_ofMethod and apparatus for lightweight and parallelization of accelerator task scheduling