5 papers
Moebius: Serving Mixture-of-Expert Models with Seamless Runtime Parallelism Switch
Shaoyu Wang, Yizhuo Liang, Jaeyong Song +2
Mixture-of-Experts (MoE) architectures scale large language models (LLMs) to hundreds of billions of parameters. Serving a single MoE model requires multiple GPUs operating in para…
Cornserve: A Distributed Serving System for Any-to-Any Multimodal Models
Jae-Won Chung, Jeff J. Ma, Jisang Ahn +4
Any-to-Any models are an emerging class of multimodal models that accept combinations of multimodal data (e.g., text, image, video, audio) as input and generate them as output. Ser…
Cornfigurator: Automated Planning for Any-to-Any Multimodal Model Serving
Jeff J. Ma, Jae-Won Chung, Jisang Ahn +5
Any-to-Any models are an emerging class of multimodal models that accept combinations of text and multimodal data as input and generate them as output, introducing heterogeneous co…
Skeleton-Guided Diffusion Model for Accurate Foot X-ray Synthesis in Hallux Valgus Diagnosis
Midi Wan, Pengfei Li, Yizhuo Liang +4
Medical image synthesis plays a crucial role in providing anatomically accurate images for diagnosis and treatment. Hallux valgus, which affects approximately 19% of the global pop…
Back to Fundamentals: Low-Level Visual Features Guided Progressive Token Pruning
Yuanbing Ouyang, Yizhuo Liang, Qingpeng Li +5
Vision Transformers (ViTs) excel in semantic segmentation but demand significant computation, posing challenges for deployment on resource-constrained devices. Existing token pruni…