Showing 2026Show all
2 papers · 1 filter
cs.AR2026
NeuroFlex: Lossless Element-Level ANN-SNN Co-Execution for Efficient Sparse Inference
Varun Manjunath, Pranav Ramesh, Gopalakrishnan Srinivasan
Sparse DNN accelerators specialize in ANN or SNN execution, leaving energy or latency on the table when workload characteristics vary within a layer. Hybrid accelerator designs tha…
cs.AR2026
BRIM: Workload-Balanced Dual-Sided Bit-Serial Sparse Inference Accelerator
Varun Manjunath, Ruokai Yin, Donghyun Lee +2
Bit-serial accelerators exploit bit-level sparsity to reduce DNN inference cost, but existing designs exploit sparsity on only one operand, bounding the speedup. Extending sparsity…