4 papers
When the Strongest Teacher Is Not the Best Teacher: Student-Centric Answer Selection
Zhengyu Hu, Zheyuan Xiao, Linxin Song +10
LLM training increasingly relies on teacher-generated supervision, from synthetic responses to reasoning traces and tool-use demonstrations. Current practice often chooses the high…
Population-Aligned Persona Generation for LLM-based Social Simulation
Zhengyu Hu, Jianxun Lian, Zheyuan Xiao +7
Recent advances in large language models (LLMs) have enabled human-like social simulations at unprecedented scale and fidelity, offering new opportunities for computational social…
Unveiling the Learning Mind of Language Models: A Cognitive Framework and Empirical Study
Zhengyu Hu, Jianxun Lian, Zheyuan Xiao +5
Large language models (LLMs) have shown impressive capabilities across tasks such as mathematics, coding, and reasoning, yet their learning ability, which is crucial for adapting t…
Explaining Length Bias in LLM-Based Preference Evaluations
Zhengyu Hu, Linxin Song, Jieyu Zhang +7
The use of large language models (LLMs) as judges, particularly in preference comparisons, has become widespread, but this reveals a notable bias towards longer responses, undermin…