| |
 | | Date: | 30 August 2024, Friday | | Time: | 3:45pm to 4:45pm | | Venue: | School of Computing & Information Systems 1 (SCIS1), Level 2, Seminar Room 2-4 Singapore Management University, 80 Stamford Road, Singapore 178902 |
|
Limited seating. Registration will close on 25 August 2024 or once maximum capacity is reached. Registration is required for attendance. Light refreshment will be provided after the talks. | Research Cluster: Artificial Intelligence & Data Science | | | | Topic: | Towards LLM-based Fact Checking under Limited Resources | | Speaker: | GAO Wei, Associate Professor of Computer Science | | Abstract: | Automating Fact Checking (FC) is difficult as it requires human-like understanding, reasoning, and writing skills, such as comprehension of context, knowledge of common sense and facts, ability to gather evidence, and understanding of cause and effect. Recent advances in large pre-trained language models, namely LLMs, have pushed context understanding and text generation to a new level due to their transformer-based architectures, enabling them to capture subtle semantic nuances and relationships in the text. Such advantages have the potential to empower FC. However, leveraging LLMs for FC is constrained by limited data and computing resources. In this talk, I will discuss the relevant issues and solutions to tackle these challenges within an established FC pipeline including early detection of rumors, evidence retrieval, fact verification, and justification generation. |
|
| | | Research Cluster: Human-Machine Collaborative Systems | | | | Topic: | Towards Human-AI Collaborative Decision-Making for Physical Stroke Rehabilitation Assessment | | Speaker: | LEE Min Hun, Assistant Professor of Computer Science | | Abstract: | Rapid advances in artificial intelligence (AI) and machine learning (ML) have made them applicable to support healthcare practices. However, the deployment of these AI systems remains a challenge. In this talk, I will present the findings from ongoing studies to design, develop, and evaluate a human-AI collaborative decision support system for physical stroke rehabilitation assessment. |
|
| | | Research Cluster: Information Systems & Technology | | | | Topic: | Neural Controller Synthesis and Verification with Guarantees | | Speaker: | Djordje ZIKELIC, Assistant Professor of Computer Science | | Abstract: | Learning-based methods such as reinforcement learning are receiving increasing attention for solving challenging control tasks. However, the lack of safety assurances about learned controllers poses a significant barrier to their practical deployment. In this talk, we will present a framework for learning and/or formally verifying neural controllers. The framework is applicable to stochastic dynamical systems, thus also taking into account environment uncertainty. Given a property and a probability bound, the framework jointly learns and formally verifies a controller together with a formal certificate of the property being satisfied with at least the specified probability, both parametrized as neural networks. Certificates are martingale-like objects that can be effectively used to formally reason about stochastic systems in a fully automated fashion. The framework is applicable to solving reachability, safety, reach-avoidance and stability tasks, as well as logical compositions of these tasks. |
| | | | | | ABOUT THE SPEAKER(S) |  | Wei Gao is an Associate Professor in the School of Computing and Information Systems at SMU. He is generally working on the intersection of natural language processing, information retrieval and social computing, with a specific focus on information credibility and factuality in recent years. His main research findings lie in several key areas, such as misinformation detection and automated fact checking, social sentiment quantification, cross-media topic discovery and fusion, and multilingual web search. Prior to Singapore, he was a senior lecturer in Victoria University of Wellington in New Zealand, a research scientist in Qatar Computing Research Institute, and a research assistant professor in the Chinese University of Hong Kong. | | | |  | Min Lee is an Assistant Professor in the School of Computing and Information Systems at Singapore Management University. His research investigates to design, develop, and evaluate human-centered AI systems to improve healthcare practices (e.g. physical stroke rehabilitation).
| | | |  | Djordje Zikelic is an Assistant Professor at the School of Computing and Information Systems, Singapore Management University. Prior to joining SMU, he obtained his PhD in computer science at the Institute of Science and Technology Austria (ISTA) and bachelor’s and master’s degrees in mathematics at the University of Cambridge. His research focuses on formal verification and synthesis of software and intelligent systems and lies at the interplay of formal methods, trustworthy AI and programming languages research. His work is regularly published at premier venues in formal methods, programming languages, AI and ML, including CAV, FM, PLDI, POPL, OOPSLA, AAAI, NeurIPS and IJCAI. For his PhD work, he received the Outstanding PhD Thesis award and the Outstanding Scientific Achievement award at ISTA. | | | | | | | | SEMINAR MODERATOR | | | |  | TAN Ah Hwee
Lee Kong Chian Professor of Computer Science
Associate Dean (Research)
|
|
|
|
|
