AHASD is a new asynchronous heterogeneous architecture for mobile NPU-PIM systems that enables efficient adaptive speculative decoding for LLMs by decoupling drafting and verification with specialized controls and hardware units, delivering up to 4.2x throughput and 5.6x energy efficiency gains.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
AHASD: Asynchronous Heterogeneous Architecture for LLM Adaptive Drafting Speculative Decoding on Mobile Devices
AHASD is a new asynchronous heterogeneous architecture for mobile NPU-PIM systems that enables efficient adaptive speculative decoding for LLMs by decoupling drafting and verification with specialized controls and hardware units, delivering up to 4.2x throughput and 5.6x energy efficiency gains.