person · community record
Zhuohan Li
A connected record in the developer community graph.
Checking node ownership…
01
Connections
26 relationships
authoredAlpa: Automating Inter- and Intra-Operator Parallelism for Distributed Deep Learningpaper ↗authoredAlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Servingpaper ↗authoredAn Evaluation of Memory Optimization Methods for Training Neural Networkspaper ↗authoredEfficient Memory Management for Large Language Model Serving with PagedAttentionpaper ↗authoredFairness in Serving Large Language Modelspaper ↗authoredFast Structured Decoding for Sequence Modelspaper ↗authoredHigh-throughput Generative Inference of Large Language Models with a Single GPUpaper ↗authoredHint-Based Training for Non-Autoregressive Machine Translationpaper ↗authoredHoplite: efficient and fault-tolerant collective communication for task-based distributed systemspaper ↗authoredJenga: Effective Memory Management for Serving LLM with Heterogeneitypaper ↗authoredJudging LLM-as-a-judge with MT-Bench and Chatbot Arenapaper ↗authoredLMSYS-Chat-1M: A Large-Scale Real-World LLM Conversation Datasetpaper ↗authoredOn Optimizing the Communication of Model Parallelism