Wanyun Cui

Wanyun Cui

Associate Professor, Shanghai University of Finance and Economics

BIO

Wanyun Cui is an associate professor at the School of Computing and Artificial Intelligence, Shanghai University of Finance and Economics. He leads the SCALE Lab (SUFE Cognitive AI & Language Exploration Lab) here. His research interests lie at the large language model inference. He has published 20+ papers in NeurIPS, ICLR, SIGMOD, PVLDB, IJCAI, AAAI, ACL, and EMNLP. He has been recognized as an AI 2000 Most Influential Scholar Honorable Mention (2012-2021, 2013-2022) and has won the ACM China Outstanding Doctoral Dissertation Nomination Award (top 4 in China) and the ACM Shanghai Outstanding Doctoral Dissertation Award (top 2 in Shanghai).

Current Research Projects

  • LLM-based Complex Reasoning We explore techniques to enhance LLMs’ reasoning capabilities.

  • Efficient LLM Inference We optimize the efficiency of LLM inference, including long-context LLM and parameter quantization.

  • LLMs for Domain Applications (e.g. Finance and Education) We develop specialized LLM applications based on FinChat, such as automated financial analysis, research report generation, retrieval-augmented QA, and expert role-playing.

🔴 For students interested in my research: I am recruiting students who are passionate about artificial (general) intelligence and possess strong programming skills (experience with open-source projects is a plus).

What we offer:

  • Learning: Customized LLM training plans tailored to your development. And GPU resources
  • Research: Opportunities to explore cutting-edge LLM problems and publish in top AI conferences
  • Practice: Hands-on experience with real LLM applications and potential internships from leading tech companies (Alibaba, Ant Group, ByteDance, etc.)

🔴 If you are interested, please do not hesitate to contact me.

🔥 News 🔥

  1. 2025/10. I have been selected as a NeurIPS 2025 “Reviewer: Top Reviewer”.
  2. 2025/09. One paper on large language model long-context inference accepted to NeurIPS 2025.
  3. 2024/09. One paper on large language model quantization accepted to NeurIPS 2024.
  4. 2024/09. One paper on instruction generation by large language models accepted to EMNLP 2024 Findings.
  5. 2024/03. We release the Technology and Security White Paper on the Application of Large Language model in Finance with Ant Group and BCTC. pdf
  6. 2023/12. One paper accepted to AAAI 2024.
  7. 2023/07. We released the large language model FinChat, which outperforms ChatGPT (3.5) on C-Eval. Congrats to our research team! FinChat intro
  8. 2023/06. Are you interested in training your own large model chatbot? Check out the dataset collection we have gathered: LLMDataHub
  9. One paper accepted to ACL 2023.
  10. One paper accepted to ACL Findings 2023.
  11. One paper accepted to NeurIPS 2022.
  12. I was recognized as the AI 2000 Most Influential Scholars Honorable Mention (2012-20221) by AMiner, rank 31 in the database field.

📝 PUBLICATIONS

Preprint

  1. Shuyang Cai, Wanyun Cui, Evade ChatGPT Detectors via A Single Space, paper

Published

  1. Wanyun Cui, Mingwei Xu, Homogeneous Keys, Heterogeneous Values: Exploiting Local KV Cache Asymmetry for Long-Context LLMs, (NeurIPS 2025, CCF A)
  2. Wanyun Cui, Qianle Wang, Cherry on Top: Parameter Heterogeneity and Quantization in Large Language Models, (NeurIPS 2024, CCF A)
  3. Wanyun Cui, Qianle Wang, Ada-Instruct: Adapting Instruction Generators for Complex Reasoning, (EMNLP Findings 2024, CCF B)
  4. Wanyun Cui, Linqiu Zhang, Modeling Knowledge Graphs with Composite Reasoning, (AAAI 2024, CCF A)
  5. Wanyun Cui, Xingran Chen, Free Lunch for Efficient Textual Commonsense Integration in Language Models, (ACL 2023, CCF A)
  6. Wanyun Cui, Xingran Chen, Exploring Automatically Perturbed Natural Language Explanations in Relation Extraction, (ACL Findings 2023, CCF A)
  7. Wanyun Cui, Xingran Chen, Instance-based Learning for Knowledge Base Completion, (NeurIPS 2022, CCF A) paper code 中文版
  8. Wanyun Cui, Xingran Chen, Enhancing Natural Language Representation with Large-Scale Out-of-Domain Commonsense (ACL Findings 2022, CCF A) paper code
  9. Wanyun Cui, Xingran Chen, Open Rule Induction, (NeurIPS 2021, CCF A) paper code slides
  10. Wanyun Cui, Sen Yan, Isotonic Data Augmentation for Knowledge Distillation, (IJCAI 2021, CCF A) paper code
  11. Wanyun Cui, Guangyu Zheng, Wei Wang, Zero-shot domain adaptation for natural language inference by projecting superficial words out, (Knowledge-Based Systems 2021, JCR Q1)
  12. Wanyun Cui, Guangyu Zheng, Wei Wang, Unsupervised Natural Language Inference via Decoupled Multimodal Contrastive Learning, (EMNLP 2020, CCF B, oral) paper code
  13. Wanyun Cui, Guangyu Zheng, Zhiqiang Shen, Sihang Jiang, Wei Wang, Transfer Learning for Sequences via Learning to Collocate, (ICLR 2019) paper code
  14. Wanyun Cui, Yanghua Xiao, Haixun Wang, Yangqiu Song, Seung-won Hwang, Wei Wang, KBQA: Learning Question Answering over QA Corpora and Knowledge Bases, (PVLDB 2017, CCF A)
  15. Wanyun Cui, Yanghua Xiao, Wei Wang, KBQA: An Online Template Based Question Answering System over Freebase, (IJCAI 2016, CCF A), demo
  16. Wanyun Cui, Xiyou Zhou, Hangyu Lin, Yanghua Xiao, Haixun Wang, Seung-won Hwang, Wei Wang, Verb Pattern: A Probabilistic Semantic Representation on Verbs, (AAAI 2016, CCF A)
  17. Wanyun Cui, Yanghua Xiao, Haixun Wang, Wei Wang, Local Search of Communities in Large Graphs, (SIGMOD 2014, CCF A)
  18. Wanyun Cui, Yanghua Xiao, Haixun Wang, Yiqi Lu, and Wei Wang. Online Search of Overlapping Communities, (SIGMOD 2013, CCF A)
  19. Wanyun Cui, Yanghua Xiao, Chapter 14: Question Answering over Knowledge Graphs, Knowledge Graph: Concepts and Techniques (book chapter).
  20. Bo Xu, Yong Xu, Jiaqing Liang, Chenhao Xie, Bin Liang, Wanyun Cui and Yanghua Xiao, CN-DBpedia: A Never-Ending Chinese Knowledge Extraction System, IEA/AIE 2017.
  21. Yanghua xiao, Ji Hong, Wanyun Cui, Zhenying He, Wei Wang, Guodong Feng, Branch Code: An Efficient Labeling Scheme for Query Answering on Trees, (ICDE 2012, CCF A)

🎓 ALUMNI

  • Qianle Wang, ByteDance
  • Jian Sun, Baidu
  • Junhao Zhao, Huawei
  • Xingran Chen, University of Michigan, master
  • Sen Yan, University of Colorado Boulder, PhD
  • Guangyu Zheng, Ant Group
  • Le Wang, Ant Group
  • Lingyu Cai, Alibaba

🏆 AWARDS

  1. 2022 The AI 2000 Most Influential Scholar Annual List by AMiner, rank 31 in the database field.
  2. 2017 ACM China Outstanding Doctoral Dissertation Nomination Award, top 4 in China
  3. 2017 ACM Shanghai Outstanding Doctoral Dissertation Award, top 2 in Shanghai
  4. 2017 Fudan Academic Star
  5. 2013 Fudan Outstanding Undergraduate Award, top 10 for Fudan undergraduates.
  6. 2011, 2009 ACM/ICPC Beijing/Shanghai Region Invitation Contest double gold medals