2 papers
cs.AR2026
SegFold: Accelerating Sparse GEMM with a Fine-Grained Dynamic Dataflow
Xinrui Wu, Hanyu Wang, Jason Cong +1
Generalized sparse matrix-matrix multiplication (SpGEMM) is critical in many domains. Existing CPUs and GPUs, as well as specialized accelerators, rely on static dataflows (e.g., i…
cs.LG2026
TRAM: Training Approximate Multiplier Structures for Low-Power AI Accelerators
Chang Meng, Hanyu Wang, Yuyang Ye +3
Reducing power consumption in AI accelerators is increasingly important. Approximate computing can reduce power consumption while keeping the accuracy loss small. Since multipliers…