3 papers
cs.AI2026
Which Tokens Should SFT Actually Learn? A Token-Trimming Perspective on Mathematical Reasoning
Yaning Jia, Chunhui Zhang, Wenxuan Xu +3
Supervised fine-tuning (SFT) applies a uniform cross-entropy loss to all target tokens, even though different tokens provide unequal learning signals for mathematical reasoning. Th…
cs.LG2025
What Makes a Good Curriculum? Disentangling the Effects of Data Ordering on LLM Mathematical Reasoning
Yaning Jia, Chunhui Zhang, Xingjian Diao +4
Curriculum learning (CL) - ordering training data from easy to hard - has become a popular strategy for improving reasoning in large language models (LLMs). Yet prior work employs…
cs.LG2025
Scaled Supervision is an Implicit Lipschitz Regularizer
Zhongyu Ouyang, Chunhui Zhang, Yaning Jia +1
In modern social media, recommender systems (RecSys) rely on the click-through rate (CTR) as the standard metric to evaluate user engagement. CTR prediction is traditionally framed…