|
Sitao Cheng (
程思陶
)
|
Fields of Interest
My research mainly focuses on how to use language models (LMs) to reasoning over structured environments (e.g. knowledge base, tables, DBs). My interests include multi-hop reasoning (e.g. question answering), LLM-agents and knowledge graph. Currently, I am focusing on the following research topics:- LLMs interaction with structured environments. (e.g. KB reasoning by interaction.)
- Retrieval Augmented Generation. (e.g. retrieval of structured information by LLMs and LLM reasoning augmentation.)
- Semantic parsing. (Translate a natural language utterance to executable queries for precise reasoning on environments.)
- Grounded language models. (Ground a plan to structured environments.)
Other Interests:
- Body building (power lifting, crossfit, nutritions, etc)
- Basketball
Publications
[1] Call Me When Necessary: LLMs can Efficiently and Faithfully Reason over Structured Environments
The 62nd Annual Meeting of the Association for Computational Linguistics(ACL), in Bangkok, Thailand, 2024. (Pre-prints)
Sitao Cheng, Ziyuan Zhuang, Yong Xu, Fangkai Yang, Chaoyun Zhang, Xiaoting Qin, Xiang Huang, Ling Chen, Qingwei Lin, Dongmei Zhang, Saravan Rajmohan, Qi Zhang
[Arxiv]
[2] QueryAgent: A Reliable and Efficient Reasoning Framework with Environmental Feedback based Self-Correction
The 62nd Annual Meeting of the Association for Computational Linguistics(ACL), in Bangkok, Thailand, 2024. (Pre-prints)
Xiang Huang*, Sitao Cheng*, Shanshan Huang, Jiayu Shen, Yong Xu, Chaoyun Zhang, Yuzhong Qu
[Arxiv]
[3] MarkQA: A large scale KBQA dataset with numerical reasoning
The 2023 Conference on Empirical Methods in Natural Language Processing(EMNLP), in Singapore, 2023.
Xiang Huang, Sitao Cheng, Yuheng Bao, Shanshan Huang, Yuzhong Qu
[Arxiv][Code][Homepage]
[4] Question Decomposition Tree for Answering Complex Questions over Knowledge Bases
Thirty-Seventh AAAI Conference on Artificial Intelligence(AAAI), in Washington, DC, USA, 2023.
Xiang Huang, Sitao Cheng, Yiheng Shu, Yuheng Bao, Yuzhong Qu
[Arxiv][Code]
Research Experience
- 2023.10-Now Research Intern at Microsoft DKI(Data, Knowledge, Intelligence) Lab. Adviser: Yong Xu
- 2021.9-Now M.S. at Websoft Lab, Nanjing University. Adviser: Yuzhong Qu
- 2023.6-2023.10 NLP Intern at Ant Group Digitalization Management Group.
Education
- 2021.09-Now Grad Student in Institute of Computer Science, Nanjing University. Adviser: Prof. Yuzhong Qu
- 2017.09-2021.06 B.E. in School of Information and Software Engineering, University of Electronic Science and Technology of China. Adviser: Prof. Yi Ding