首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Junyu Chen · Jungang Li · Jing Xiong · Wenjie Wang · Qingyao Yang · He Xiao · Zhen Li · Taiqiang Wu · Mengzhao Chen · Zhen Peng · Chaofan Tao · Long Shi · Hongxia Yang · Ngai Wong
Taiqiang Wu · Runming Yang · Tao Liu · Jiahao Wang · Ngai Wong
Model merging, typically on Instruct and Thinking models, has shown remarkable performance for efficient reasoning. In this paper, we systematically revisit the simplest merging method that interpolates two weights directly. Particularly, we observe that model interpolation follows a three-stage...
Tao Liu · Taiqiang Wu · Runming Yang · Shaoning Sun · Junjie Wang · Yujiu Yang
Supervised fine-tuning (SFT) is a fundamental post-training strategy to align Large Language Models (LLMs) with human intent. However, traditional SFT often ignores the \textit{one-to-many} nature of language by forcing alignment with a single reference answer, leading to the model overfitting to...