MxGLUT introduces a reconfigurable LUT-centric broadcast dataflow accelerator with mixed-precision LUT-based PEs that unifies FP8-INT4 and FP8-FP8 GEMM without separate FP datapaths, reporting up to 2.16x prefill speedup and 0.492 TFLOPS/mm² area efficiency in 28nm synthesis.
Pointer sentinel mixture models
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
Prism optimizes expert placement and uses runtime migration for distributed MoE inference on heterogeneous edge GPUs, achieving up to 30.6% lower latency than baselines.
citing papers explorer
-
MxGLUT: A Reconfigurable LUT-Centric Broadcast Dataflow Accelerator for Mixed-Precision GEMM
MxGLUT introduces a reconfigurable LUT-centric broadcast dataflow accelerator with mixed-precision LUT-based PEs that unifies FP8-INT4 and FP8-FP8 GEMM without separate FP datapaths, reporting up to 2.16x prefill speedup and 0.492 TFLOPS/mm² area efficiency in 28nm synthesis.
-
Accelerating Edge Inference for Distributed MoE Models with Latency-Optimized Expert Placement
Prism optimizes expert placement and uses runtime migration for distributed MoE inference on heterogeneous edge GPUs, achieving up to 30.6% lower latency than baselines.