8 papers · 1 filter
SPRI: SVD-Partitioned Residual Initialization for Data-Constrained MoE Upcycling
Weiqiao Shan, Ruixiang Mao, Yuang Li +10
Mixture-of-Experts (MoE) models enable efficient scaling, but training them from scratch remains prohibitively expensive. MoE upcycling mitigates this cost by converting pretrained…
Taylor-Calibrate: Principled Initialization for Hybrid Linear Attention Distillation
Zhongzhu Zhou, Qingyang Wu, Junxiong Wang +4
Hybrid linear attention models offer an appealing path to faster long-context inference: they reduce the quadratic cost and KV-cache burden of full softmax attention while retainin…
NLI:Non-uniform Linear Interpolation Approximation of Nonlinear Operations for Efficient LLMs Inference
Jiangyong Yu, Xiaomeng Han, Xing Hu +3
Large Language Models (LLMs) have demonstrated remarkable performance across a wide range of tasks, but their deployment is often constrained by substantial memory footprints and c…
MoEQuant: Enhancing Quantization for Mixture-of-Experts Large Language Models via Expert-Balanced Sampling and Affinity Guidance
Xing Hu, Zhixuan Chen, Dawei Yang +5
Mixture-of-Experts (MoE) large language models (LLMs), which leverage dynamic routing and sparse activation to enhance efficiency and scalability, have achieved higher performance…
RWKVQuant: Quantizing the RWKV Family with Proxy Guided Hybrid of Scalar and Vector Quantization
Chen Xu, Yuxuan Yue, Zukang Xu +6
RWKV is a modern RNN architecture with comparable performance to Transformer, but still faces challenges when deployed to resource-constrained devices. Post Training Quantization (…
MambaQuant: Quantizing the Mamba Family with Variance Aligned Rotation Methods
Zukang Xu, Yuxuan Yue, Xing Hu +7
Mamba is an efficient sequence model that rivals Transformers and demonstrates significant potential as a foundational architecture for various tasks. Quantization is commonly used…