Zhenya Huang

Perception-R1: Advancing Multimodal Reasoning Capabilities of MLLMs via Visual Perception Reward

Not All Code Helps: Disentangling the Impact of Code Data on Mathematical Reasoning in Large Language Models

Pruning Long Chain-of-Thought of Large Reasoning Models via Small-Scale Preference Optimization

Find tailored step example for next step: A Targeted Step-wise Retrieval Framework for Guiding LLM Reasoning

Nested Hash Layer: A Plug-and-play Module for Multiple-length Hash Code Learning

MARCOS: Deep Thinking by Markov Chain of Continuous Thoughts

Auditing Test Data Contamination with Error Rate Control for Reliable LLM Evaluation

What Makes In-context Learning Effective for Mathematical Reasoning

CogMath: Assessing LLMs' Authentic Mathematical Ability from a Human Cognitive Perspective

A Closed-Form Solution for Fast and Reliable Adaptive Testing

Unveiling the Magic of Code Reasoning through Hypothesis Decomposition and Amendment

Can LLMs Solve Longer Math Word Problems Better?

TestAgent: An Adaptive and Intelligent Expert for Human Assessment

CogMath: Evaluating LLMs' Authentic Mathematical Ability from a Cognitive Perspective

Self-Informed Generative Active Learning

Beyond Layers: A Global Message-Passing Mechanism for Heterophilic Graphs