5 papers
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…
SHAPE: Coalition-Aware Expert Pruning for Sparse Mixture-of-Experts LLMs
Yuhao Zhang
Sparse Mixture-of-Experts (MoE) large language models achieve strong quality with low per-token compute, yet their deployment is often limited by the memory wall: the full expert p…
Enhancing Speech Large Language Models with Prompt-Aware Mixture of Audio Encoders
Weiqiao Shan, Yuang Li, Yuhao Zhang +9
Connecting audio encoders with large language models (LLMs) allows the LLM to perform various audio understanding tasks, such as automatic speech recognition (ASR) and audio captio…
Leveraging Unit Language Guidance to Advance Speech Modeling in Textless Speech-to-Speech Translation
Yuhao Zhang, Xiangnan Ma, Kaiqi Kou +7
The success of building textless speech-to-speech translation (S2ST) models has attracted much attention. However, S2ST still faces two main challenges: 1) extracting linguistic fe…
Optimizing Speech Multi-View Feature Fusion through Conditional Computation
Weiqiao Shan, Yuhao Zhang, Yuchen Han +7
Recent advancements have highlighted the efficacy of self-supervised learning (SSL) features in various speech-related tasks, providing lightweight and versatile multi-view speech…