8 papers
Cassandra: Enabling Reasoning LLMs at Edge via Self-Speculative Decoding
Soongyu Choi, Yuntae Kim, Muyoung Son +1
Speculative decoding has emerged as a promising lossless approach for accelerating Large Language Models (LLMs). As reasoning LLMs increasingly suffer from decode-stage overhead an…
DiSC: Resolution-Scalable Acceleration of Diffusion Models by Exploiting Sparsity and Cached Token Reuse with Hash-based Distribution
Jieon Yoon, Hangyeol Lee, Jaehoon Heo +1
Transformer-based diffusion models offer superior scalability and performance but suffer from high computational overhead due to the iterative nature and quadratic complexity of se…
MASQ: Accelerating Masked Diffusion via Stage-Wise Multi-Precision Quantization
Seeyeon Kim, Jaehun Lee, Sungyeob Yoo +1
Masked diffusion enables region-specific image synthesis but suffers from computational redundancy, since the entire image is processed each timestep even though only the masked re…
ORBIS: Output-Guided Token Reduction with Distribution-Aware Matching for Video Diffusion Acceleration
Hangyeol Lee, Joo-Young Kim
Diffusion Transformer (DiT) has emerged as a powerful model architecture for generating high-quality images and videos. In the case of video DiT, 3D Spatio-Temporal Attention incre…
Rethinking Token Reduction for Diffusion Models via Output-Similarity-Awareness
Hangyeol Lee, Hyojeong Lee, Joo-Young Kim
Diffusion Transformers (DiTs) achieve superior image generation quality but suffer from quadratic computational complexity relative to token count. While various token reduction (T…
CoX-MoE: Coalesced Expert Execution for High-Throughput MoE Inference with AMX-Enabled CPU-GPU Co-Execution
Muyoung Son, Yi Chen, Seungjae Yoo +2
The Mixture-of-Experts (MoE) architecture improves computational efficiency via sparse expert activation, but throughput-oriented inference faces substantial GPU memory pressure du…