5 papers
Temporal-Attention Head Specialization During Video Diffusion Training
Taewoo Ha, Shafayat Mowla Anik, Dae Yeol Lee +2
Video diffusion transformers depend on temporal attention to coordinate information across frames, yet nearly everything known about this mechanism comes from analyzing trained mod…
Lesion-DDPM: Lesion-Enhanced 3D Diffusion for MS MRI Synthesis
Weidong Zhang, Yongchan Jung, Shafayat Mowla Anik +6
3D FLAIR MRI is widely recommended as one of the standard MRI sequences for brain imaging in multiple sclerosis (MS), but publicly available MS datasets remain relatively small and…
RH+: Row-Hit-Optimized Scheduling for PIM-based LLM Inference
Yongchan Jung, Shafayat Mowla Anik, Byeong Kil Lee +1
Large language model inference on processing-in-memory (PIM) architectures promises to break the memory wall by performing multiply-accumulate (MAC) operations directly within HBM3…
Activation Concentration: Characterizing Column-Level Output Sparsity Across Diffusion Model Architectures
Dazhi Yang, Shafayat Mowla Anik, Byeong Kil Lee +1
Recent diffusion accelerators exploit activation sparsity by skipping near-zero GELU outputs, reporting 52--85% element-level sparsity. However, systolic-array hardware processes a…
Dead on Arrival: Characterizing and Protecting Against Dead-Entry TLB Misses in GPU Microarchitectures
Shafayat Mowla Anik, Yongchan Jung, Jeeho Ryoo +1
GPU workloads with large memory footprints frequently suffer from redundant L2 TLB misses in which a recently evicted translation is immediately re-walked at full page-walk cost. W…