Ziyu Shang

I am a second-year PhD at Southeast University advised by Prof. Peng Wang and Assoc. Prof. Wenjun Ke . I completed my undergraduate degree in School of Computer Science and Engineering at NJUST.

I'm interested in building a trustworthy AI system that can assess its own factual knowledge, learn new information systematically, and intelligently allocate time, compute, and storage resources to handle complex tasks with enhanced factual accuracy and reasoning capabilities.

Google Scholar  /  GitHub  /  Email

• 2025-06 Learning Multi-Granularity and Adaptive Representation for Knowledge Graph Reasoning is accepted by IEEE Transactions on Knowledge and Data Engineering (TKDE).
• 2025-05 Two papers accepted as ACL 2025 Oral (Top 8%), focusing on mechanisms of novel knowledge learning in large language models and topic modeling in the era of large language models.
• 2024-09 Unveiling factuality and injecting knowledge for LLMs via reinforcement learning and data proportion is accepted by Science China Information Sciences (SCIS).
• 2024-04 Learning Multi-Granularity and Adaptive Representation for Knowledge Graph Reasoning is accepted by IJCAI 2024.
• 2024-03 OntoFact: Unveiling Fantastic Fact-Skeleton of LLMs via Ontology-Driven Reinforcement Learning is accepted by AAAI 2024 Oral (Top 2.3%).
• 2023-07 ASKRL is accepted by ISWC 2023.

* denotes equal contribution