Lili Mou
Appointment
Canada CIFAR AI Chair
Pan-Canadian AI Strategy
About
Appointed Canada CIFAR AI Chair – 2020
Lili Mou is a Canada CIFAR AI Chair and a fellow at Amii. He is appointed as an assistant professor in the Department of Computing Science at the University of Alberta.
Mou’s is interested in applying deep learning to natural language processing. He seeks to build an intelligent system that can understand and interact with humans via natural language, involving both text understanding and text generation. He focuses on fundamental problems in machine learning and deep learning methods applied to natural language processing. Towards this goal, Mou focuses on fundamental machine learning problems, such as neuro-symbolic reasoning, discrete search-based text generation, and ensemble-based multi-teacher distillation.
Awards
- Global Top Chinese Young Scholars in Artificial Intelligence, Baidu Inc., 2022
- AAAI New Faculty Highlight Award, 2021
- Best Paper Nomination, Association for Computational Linguistics, 2019
- Distinguished Ph.D Thesis Award
Relevant Publications
- Behzad Shayegh, Yuqiao Wen, Lili Mou. (2024). Ensemble-Based Unsupervised Discontinuous Constituency Parsing by Tree Averaging. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL).
- Yongchang Hao, Yanshuai Cao, Lili Mou. (2024). Flora: Low-Rank Adapters Are Secretly Gradient Compressors. In Proceedings of the 41st International Conference on Machine Learning (ICML).
- Wen, Y., Hao, Y., Cao, Y., & Mou, L. (2023). An Equal-Size Hard EM Algorithm for Diverse Dialogue Generation. In International Conference on Learning Representations (ICLR).
- Liu, P., Huang, C., & Mou, L. (2022). Learning Non-Autoregressive Models from Search for Unsupervised Sentence Summarization. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (ACL).
Li, J., Li, Z., Mou, L., Jiang, X., Lyu, M. R., & King, I. (2020). Unsupervised text generation by learning from search. In Advances in Neural Information Processing System (NeurIPS), 2020.