EdgeMM is a 22nm multi-core CPU that combines systolic-array and compute-in-memory coprocessors with adaptive activation-aware pruning, reported as 2.84x faster than a laptop RTX 3060 for multimodal LLMs.
22.9 a 12nm 18.1tflops/w sparse transformer processor with entropy-based early exit, mixed-precision predication and fine- grained power management,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
EdgeMM: Multi-Core CPU with Heterogeneous AI-Extension and Activation-aware Weight Pruning for Multimodal LLMs at Edge
EdgeMM is a 22nm multi-core CPU that combines systolic-array and compute-in-memory coprocessors with adaptive activation-aware pruning, reported as 2.84x faster than a laptop RTX 3060 for multimodal LLMs.