Fast Video Generation with Sliding Tile Attention
Paper • 2502.04507 • Published • 51
AI & ML interests
We are passionate about designing strong, efficient, and secure machine learning models and algorithms, and in building scalable, practical distributed systems that can support real-world machine learning workloads.
zhisbug
authored a
paper about 1 year ago
zhisbug
authored 17
papers over 1 year ago
LightSeq: Sequence Level Parallelism for Distributed Training of Long Context Transformers
Paper • 2310.03294 • Published • 2
Online Speculative Decoding
Paper • 2310.07177 • Published • 3
Judging LLM-as-a-judge with MT-Bench and Chatbot Arena
Paper • 2306.05685 • Published • 40
Evaluating the Robustness of Text-to-image Diffusion Models against Real-world Attacks
Paper • 2306.13103 • Published • 2
DistServe: Disaggregating Prefill and Decoding for Goodput-optimized Large Language Model Serving
Paper • 2401.09670 • Published • 2
Break the Sequential Dependency of LLM Inference Using Lookahead Decoding
Paper • 2402.02057 • Published
Efficient Memory Management for Large Language Model Serving with PagedAttention
Paper • 2309.06180 • Published • 49
CLLMs: Consistency Large Language Models
Paper • 2403.00835 • Published • 4
Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference
Paper • 2403.04132 • Published • 40
Toward Inference-optimal Mixture-of-Expert Large Language Models
Paper • 2404.02852 • Published
Megalodon: Efficient LLM Pretraining and Inference with Unlimited Context Length
Paper • 2404.08801 • Published • 66
MPCFormer: fast, performant and private Transformer inference with MPC
Paper • 2211.01452 • Published • 1
Optimizing Speculative Decoding for Serving Large Language Models Using Goodput
Paper • 2406.14066 • Published • 3
AdaMoE: Token-Adaptive Routing with Null Experts for Mixture-of-Experts Language Models
Paper • 2406.13233 • Published • 1
Efficient LLM Scheduling by Learning to Rank
Paper • 2408.15792 • Published • 20
Specifications: The missing link to making the development of LLM systems an engineering discipline
Paper • 2412.05299 • Published • 1
Efficiently Serving LLM Reasoning Programs with Certaindex
Paper • 2412.20993 • Published • 36