Profile
PhD student at the School of Computer Science, Sichuan University, advised by Prof. Wenqiang Lei. My research focuses on natural language processing, LLM agents, and knowledge engineering, with the objective of enabling automation of complex real-world procedures and workflows. I am particularly interested in planning for complex tasks, game agents and multi-agent systems, skill acquisition for intelligent agents, and the application of structured knowledge to improve agent reasoning and robustness.
Education
- Bachelor of Science in Computer Science and Technology — Sichuan University, 2018–2022
- Master of Science in Computer Science and Technology — Sichuan University, 2022–2024 · Advisor: Anthony (Tony) G. Cohn
- PhD in Computer Science and Technology — Sichuan University, 2024–Present · Advisor: Wenqiang Lei
Publications
NL ⇒ Schedule: Evaluate Multitask Scheduling Capability of Large Language Models
Wenrui Liao, Weihong Du, Yi Li, Hongru Liang, and Wenqiang Lei, ACL, 2026 · PDF
Unsupervised Semantic Discovery via Global and Local Semantic Alignment in Multimodal Clustering
BAR: A Backward Reasoning based Agent for Complex Minecraft Tasks
Weihong Du, Wenrui Liao, Binyu Yan, Hongru Liang, Anthony G Cohn, Wenqiang Lei, ACL, 2025 · PDF
Paged: A benchmark for procedural graphs extraction from documents
Weihong Du, Wenrui Liao, Hongru Liang, Wenqiang Lei, ACL, 2024 · PDF
CARE: A Clue-guided Assistant for CSRs to Read User Manuals
Weihong Du, Jia Liu, Zujie Wen, Dingnan Jin, Hongru Liang, Wenqiang Lei, ACL, 2024 · PDF
Knowing-how & Knowing-that: A New Task for Machine Comprehension of User Manuals
Hongru Liang, Jia Liu, Weihong Du, Dingnan Jin, Wenqiang Lei, Zujie Wen, Jiancheng Lv, ACL, 2023 · PDF
Projects & Experience
- Key R&D Project / National Key R&D Program — Ministry of Science and Technology, Project No. 2022YFC3301503 · Research on supervision and prediction techniques for important case-handling nodes based on event chains (2022-10-01 – 2025-09-30)
- General Program — National Natural Science Foundation of China (NSFC), Project No. 62272330 · Research on conversational question answering systems for complex heterogeneous data (2023-01-01 – 2026-12-31)
- NSFC Joint Fund — National Natural Science Foundation of China, Project No. U24A20328-LH · Research on domain-knowledge-guided multimodal large-model learning and reasoning techniques (2025-01-01 – 2028-12-31)
- Young Scientists Fund (Category C) — National Natural Science Foundation of China (NSFC), Project No. 62206191 · Research on music text representation learning based on multi-source information fusion (2023-01-01 – 2025-12-31)
Skills
- Agent systems & reasoning: Designing and evaluating tool-using agents, chain-of-thought and symbolic–neural hybrid reasoning
- Data & evaluation: Large-scale data curation, annotation pipelines, synthetic data generation, benchmark design and robustness evaluation
- Knowledge & graphs: Knowledge graph integration, schema design, knowledge-grounded generation and retrieval