6 papers
SocraticPO: Policy Optimization via Interactive Guidance
Zirui Liu, Tingyue Pan, Jie Ouyang +8
Reinforcement learning (RL) for large language models usually supervises reasoning with scalar outcome rewards, such as binary correctness. Such rewards provide an optimization dir…
UniCog: Uncovering Cognitive Abilities of LLMs through Latent Mind Space Analysis
Jiayu Liu, Yinhe Long, Zhenya Huang +1
A growing body of research suggests that the cognitive processes of large language models (LLMs) differ fundamentally from those of humans. However, existing interpretability metho…
Pruning Long Chain-of-Thought of Large Reasoning Models via Small-Scale Preference Optimization
Bin Hong, Jiayu Liu, Kai Zhang +3
Recent advances in Large Reasoning Models (LRMs) have demonstrated strong performance on complex tasks through long Chain-of-Thought (CoT) reasoning. However, their lengthy outputs…
CogMath: Assessing LLMs' Authentic Mathematical Ability from a Human Cognitive Perspective
Jiayu Liu, Zhenya Huang, Wei Dai +7
Although large language models (LLMs) show promise in solving complex mathematical tasks, existing evaluation paradigms rely solely on a coarse measure of overall answer accuracy,…
End-to-End Graph Flattening Method for Large Language Models
Bin Hong, Jinze Wu, Jiayu Liu +5
In recent years, the breakthrough of Large Language Models (LLMs) offers new ideas for achieving universal methods on graph data. The common practice of converting graphs into natu…
Bit-mask Robust Contrastive Knowledge Distillation for Unsupervised Semantic Hashing
Liyang He, Zhenya Huang, Jiayu Liu +4
Unsupervised semantic hashing has emerged as an indispensable technique for fast image search, which aims to convert images into binary hash codes without relying on labels. Recent…