oFFN: Outlier and Neuron-aware Structured FFN for Fast yet Accurate LLM Inference

제목
oFFN: Outlier and Neuron-aware Structured FFN for Fast yet Accurate LLM Inference
저자
이영민
발행일
2026-03-24
학회명
ARCHITECTURAL SUPPORT FOR PROGRAMMING LANGUAGES AND OPERATING SYSTEMS (ASPLOS)