I am a Research Associate at the Cambridge Faculty of Law, working with Prof. Felix Steffek on AI and Law. Previously, I worked as a Research Assistant at the Centre for Business Research, working with Prof. Felix Steffek and Prof. Simon Deakin on the Legal Systems and Artificial Intelligence project from 2022 to 2023.
I completed my PhD in Computer Science in the NLIP group at the University of Cambridge supervised by Prof. Ann Copestake. My PhD thesis focused on robust evaluation for NLG tasks for grammaticality, faithfulness and diversity. Tasks that were investigated include image captioning, sports news generation and topic shift aware dialogue systems. I jointly organised the weekly Cambridge NLIP seminars with Georgi Karadzhov in my third year. I completed the MPhil in Advanced Computer Science at the University of Cambridge. My MPhil thesis, entitled “How clever are the models exhibiting ‘super-human’ performance on the CLEVR VQA dataset?”, looked at diagnostic evaluations of multimodal VQA models using synthetic data. I did my MEng and BSc in Computer Science at Beihang University, supervised by Prof. Wei Li and Dr. Jie Luo, working on revision calculus and mathematical logic.
I previously held the Treasurer position (2022/23) and the Chair (2021/22) of Women@CL at the Department of Computer Science and Technology. In Women@CL, we have a positive action programme focusing on research, leadership and enterprise. We hope to support women in computing research in their careers and encourage them to aspire to leadership positions, both in academia and industry. In the past, I held the Cambridge chair for Oxbridge Women in Computer Science 2020.
I am also a member of Downing College. I held the position of Chief Operation Officer (COO) at Downing Enterprise for 2021/22. We organise annual pre-seed startup competitions directed by alumni of Downing College to promote enterpreneurship at the University of Cambridge.
Outside of the lab, I enjoy camping, baking, and cafe crawling. I am a huge fan of musicals and ballet dances. I also go gallery crawling from time to time. I have been organising the Cambridge PhD Multidisciplinary seminars since 2019.
News
- (04/2024) My PhD thesis, titled “Evaluating Natural Language Generation Tasks for Grammaticality, Faithfulness and Diversity”, is now online in the Cambridge Apollo Repository.
- (12/2023) I will be joining the Faculty of Law at the University of Cambridge as a Research Associate in January 2024, working on AI and Law.
- (08/2023) Our paper The Cambridge Law Corpus: A Corpus for Legal AI Research has been accepted to appear in the Datasets and Benchmarks Track at NeurIPS 2023.
Publications
- The Cambridge Law Corpus: A Corpus for Legal AI Research. Andreas Östling, Holli Sargeant, Huiyuan Xie, Ludwig Bull, Alexander Terenin, Leif Jonsson, Måns Magnusson, Felix Steffek. In Datasets and Benchmarks Track (NeurIPS 2023). [PDF]
- TIAGE: A Benchmark for Topic-Shift Aware Dialog Modeling. Huiyuan Xie, Zhenghao Liu, Chenyan Xiong, Zhiyuan Liu, and Ann Copestake. In Findings of Conference on Empirical Methods in Natural Language Processing (Findings of EMNLP 2021). [PDF]
- Going Beneath the Surface: Evaluating Image Captioning for Grammaticality, Truthfulness and Diversity. Huiyuan Xie, Tom Sherborne, Alexander Kuhnle, Ann Copestake. In Workshop on Evaluating Evaluation of AI Systems (AAAI 2020). [PDF]
- Audio Visual Scene-Aware Dialog System Using Dynamic Memory Networks. Huiyuan Xie, Ignacio Iacobacci. In Workshop on Dialog System Technology Challenge (AAAI 2020).
- Perceptual-Sensitive GAN for Generating Adversarial Patches. Aishan Liu, Xianglong Liu, Jiaxin Fan, Yuqing Ma, Anlan Zhang, Huiyuan Xie, Dacheng Tao. In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI 2019). [PDF]
- How clever is the FiLM model, and how clever can it be?. Alexander Kuhnle, Huiyuan Xie, Ann Copestake. In Workshop on Shortcomings in Vision and Language (ECCV 2018). [PDF]
- An Algorithm to Compute Minimal Unsatisfiable Subsets for a Decidable Fragment of First-Order Formulas. Huiyuan Xie, Jie Luo. In IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2016).