3 papers
cs.AR2025
EdgeMM: Multi-Core CPU with Heterogeneous AI-Extension and Activation-aware Weight Pruning for Multimodal LLMs at Edge
Kangbo Bai, Le Ye, Ru Huang +1
Emerging multimodal LLMs (MLLMs) exhibit strong cross-modality perception and reasoning capabilities and hold great potential for various applications at edge. However, MLLMs typic…
cs.CV2025
Accelerating Diffusion Transformer via Increment-Calibrated Caching with Channel-Aware Singular Value Decomposition
Zhiyuan Chen, Keyi Li, Yifan Jia +2
Diffusion transformer (DiT) models have achieved remarkable success in image generation, thanks for their exceptional generative capabilities and scalability. Nonetheless, the iter…
cs.AR2025
Leveraging Compute-in-Memory for Efficient Generative Model Inference in TPUs
Zhantong Zhu, Hongou Li, Wenjie Ren +4
With the rapid advent of generative models, efficiently deploying these models on specialized hardware has become critical. Tensor Processing Units (TPUs) are designed to accelerat…