![avatar2.jpg](/assets/img/avatar2.jpg?979379eb3bb27557e1ff6d4589365033)
Liangming Pan (潘亮铭)
Assistant Professor
College of Information Science
University of Arizona
Hey, thanks for stopping by! 👋
I am an Assistant Professor at the College of Information Science, University of Arizona. Previously, I was a postdoctoral scholar at the Natural Language Processing Group, University of California, Santa Barbara (UCSB), where I was fortunate to be advised by Prof. William Yang Wang. I completed my PhD from National University of Singapore in 2022, where I was advised by Prof. Min-Yen Kan and Prof. Tat-Seng Chua and worked on complex question answering and generation. Previously, I received my Master’s degree from Tsinghua University (2014 - 2017, advised by Prof. Juanzi Li and Prof. Jie Tang) and my Bachelor’s degree from Beihang University (2010 - 2014).
My primary research lies in the area of natural language processing and machine learning. The overarching goal of my research is: how to build trustworthy large language models that are logical, truthful, and safe. Specifically:
- Reasoning: How/why can LLMs consistently perform logical and faithful reasoning?
- Truthfulness: How can LLMs produce accurate, un-biased, and up-to-date information?
- Safety: How can we discover and mitigate the potential harm caused by LLMs?
📢 Annoncement
I am looking for highly motivated Ph.D./master/undergraduate students to work with me on exciting research projects related to large language models. If you are interested in working with me, please feel free to reach out to me via email.
News
Jul, 2024 | I will serve as an Area Chair for EMNLP 2024 and COLING 2025. |
---|---|
Jun, 2024 | Invited talk “Empowering Large Language Models with Faithful Reasoning” at Tsinghua University, Peking University, Xi’an Jiaotong University, and Harbin Institute of Technology (Shenzhen). |
May, 2024 | 6 papers accepted by ACL 2024 (2 Main Conference, 4 Findings). Topics involves logical reasoning, uncertainty estimation, and self-correction of LLMs. 2 papers accepted by ICML 2024. One paper is on understanding the chain-of-thought reasoning ability of LLMs. The other paper is on analyzing the impact of social media influencers on AI research visibility. |
Apr, 2024 | I will serve as the Student Volunteer Chair and an Area Chair of ACL 2024. |
Dec, 2023 | Invited talk “Combating Misinformation in the age of LLMs” at NUS Centre for Trusted Internet and Community. [Slides] |
Nov, 2023 | Our paper Attacking Open-domain Question Answering by Injecting Misinformation received the Area Chair Award (Question Answering) at IJCNLP-AACL 2023. |
Oct, 2023 | I have 8 papers accepted by EMNLP 2023 (4 Main Conference, 3 Findings, 1 Demo). Topics involves logical reasoning, safety, and evaluation of LLMs. |
Aug, 2023 | New Survey Paper! Automatically Correcting Large Language Models: Surveying the landscape of diverse self-correction strategies. We also create a paper list. |
May, 2023 | New Preprint! Logic-LM: Empowering Large Language Models with Symbolic Solvers for Faithful Logical Reasoning. Our paper Fact-Checking Complex Claims with Program-Guided Reasoning was accepted by the main conference of ACL 2023. |
Feb, 2023 | Our paper Hashtag-Guided Low-Resource Tweet Classification was accepted by WWW 2023. Invited Talk “Building Data-efficient and Explainable Fact-Checking Models” at Mohamed bin Zayed University of Artificial Intelligence (MBZUAI). |
Dec, 2022 | I will serve as an Area Chair for the Question Answering track of ACL 2023. Starting from Dec 2022, I am thrilled to join the UC Santa Barbara Natural Language Processing Group as a Postdoctoral Scholar, working with Prof. William Yang Wang. |
Selected Publications
- AACL / IJCNLP Oral PresentationAttacking Open-domain Question Answering by Injecting MisinformationIn International Joint Conference on Natural Language Processing and Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (IJCNLP-AACL), 2023Area Chair Award (Question Answering Track)
- AACL / IJCNLP Oral Presentation
- AACL / IJCNLP Oral Presentation
-