← Search

Wangyi Jiang

2 accepted papers

2025

Improved Sparse Upcycling for Instruction Tuning

COLING 2025main

The Mixture-of-Experts (MoE) architecture has demonstrated significant potential in both large-scale pre-training and instruction tuning by offering increased parameter capacity without additional inference costs. However, developing MoE models faces challenges including training instability and the…