Welcome to Dawei’s Homepage!
I am Dawei Li, a first-year PhD student at Arizona State University (ASU), advised by Prof. Huan Liu. Previously, I obtained my bachelor’s degree in Computer Science from Beijing Language and Cultural University (BLCU) and master’s degree in Data Science from University of California, San Diego (UCSD). I had also been a research intern at Beijing Advanced Innovation Center for Language Resources and AI Lab of Xiaomi. I am the founder of the opensource research community OracleLLM. I am very fortunately advised by and maintain closely connected with Prof. Tianlong Chen, Prof. Jingbo Shang and Dr. Yanran Li.
Research Interests
My research interests include Natural Language Processing and Text Mining. Here are some directions my previous work covered:
- Exploring the concept of LLMs-as-Oracle, which refers to replacing human-generated golden references with the output and feedback produced by large language models (LLMs), treating them as authoritative benchmarks and learning signals for various tasks and applications such as data synthesis, LLM-as-a-judge and self-feedback.
- Benchmarking and improving the group social intelligence of LLM agent with more dynamic, complex methods, finding high-level insights to optimize multi-agent pipeline designing.
- Applying machine learning and data science methods, like adversarial training and casual inference, in NLP tasks to build more robust, trustworthy and interpretable models.
Talk
- NICE x Soochow University: LLM’s for Data Annotation and Synthesis: A Survey
Service
- Serve as a reviewer for EMNLP 2023, NAACL 2024, EMNLP 2024, ICLR 2025, AISTATS 2025 and NeurIPS 2024 Workshop MATH-AI
- Serve as a subreviewer for CIKM 2024
- Serve as a regular reviewer for ARR
Experience
- University of California, San Diego, 2023.2 - 2024.6.
- Research Intern.
- Worked with Prof. Jingbo Shang on Robust Relation Extraction, Adversarial Training and Large Language Models.
- AI Lab of Xiaomi, 2021.8 - 2022.6.
- Machine Learning Engineering Intern.
- Worked with Dr. Yanran Li on Commonsense Knowledge Graph and Emotional Support Conversation.
- Beijing Advanced Innovation Center for Language Resources, 2020.6 - 2021.4.
- Research Intern.
- Worked on Multi-word Expression and Definition Generation.
Awards
- Outstanding Graduate Thesis Award (top 5%) at BLCU, 06/2022.
- Construction of Chinese Common Sense Knowledge Graph for Dialogue
- The Meritorious Prize winner (top 10%) in “MCM (Mathematical Contest in Modeling)”, 02/2021.
- Analysis of Hornets’ Propagation Tendency and Reliability Assessment of Sighting Reports
- The 1st Prize winner (top 8%) in “Beijing Mathematical Contest in Modeling”, 09/2020.
- Credit Decisions for Small and Micro Businesses
- The 1st Prize winner (provincial level, top 10%) in “Contemporary Undergraduate MCM”, 05/2020.
- Risk Analysis of COVID-19