About Me

Freda (Haoyue) SHI

Greetings! I just received my B.S. degree with summa cum laude from Peking University. I am now at ByteDance AI Lab as a research intern, working with my excellent mentors Dr. Hao Zhou and Dr. Lei Li . I will soon start my Ph.D. pursuit at Toyota Technological Institute at Chicago in October.

My research interests are computational linguistics and natural language processing. Specifically, I study semantic representations from computational, logical and multimodal perspectives of view.


    Conference and Journal Papers
  • Haoyue Shi*, Jiayuan Mao*, Tete Xiao*, Yuning Jiang and Jian Sun. 2018. Learning Visually-Grounded Semantics from Contrastive Adversarial Samples. In Proceedings of the 27th International Conference on Computational Linguistics (COLING 2018). (*: Equal Contribution) [PDF][code]

  • Haoyue Shi, Xihao Wang, Yuqi Sun and Junfeng Hu. 2018. Constructing High Quality Sense-specific Corpus and Word Embedding via Unsupervised Elimination of Pseudo Multi-sense. In Proceedings of the 11th Language Resources and Evaluation Conference (LREC 2018). [PDF]

  • Haoyue Shi, Jia Chen and Alexander G. Hauptmann. 2017. Joint Saliency Estimation and Matching using Image Regions for Geo-Localization of Online Video. In Proceedings of the 2017 ACM International Conference on Multimedia Retrieval (ICMR 2017). [PDF]

  • Haoyue Shi, Caihua Li and Junfeng Hu. 2016. Real Multi-Sense or Pseudo Multi-Sense: An Approach to Improve Word Representation. In Proceedings of the 1st Workshop on Computational Linguistics for Linguistic Complexity (CL4LC, associated with COLING 2016). [PDF]

  • Shan Xu, Haoyue Shi, Xiaohui Duan, Tiangang Zhu, Peihua Wu and Dongyue Liu. 2016. Cardiovascular Risk Prediction Method Based on Test Analysis and Data Mining Ensemble System. In Proceedings of the 2016 IEEE International Conference on Big Data Analysis (ICBDA 2016).

  • Thesis
  • Haoyue Shi, On Multi-Sense Word Embeddings via Matrix Factorization and Matrix Transformation, B.S. Thesis, Peking University School of EECS, May 2018. (in Simplified Chinese, won the best 10 Dissertation Award at Peking University) [PDF][slides]

  • Unpublished Manuscripts
  • Haoyue Shi, Yuqi Sun and Junfeng Hu. Understanding and Improving Multi-Sense Word Embeddings via Extended Robust Principal Component Analysis. 2018. ArXiv Preprint. [PDF]

Selected Projects


Implementation of our paper on COLING'18, Learning Visually-Grounded Semantics from Contrastive Adversarial Samples.

...   ...


Source code of our study on pseudo multi-sense.

...   ...


  • I was an ACM-ICPC competitor. I won two gold medals on regional contests with my teammates Chengxian Mo and Tianshi Li.

  • I play Guqin. At the very first time, I follow Jiu-Yi School. My style is also influenced by Mr. Gong Yi a lot. I served as the minister of PKU Guqin Society from August 2016 to August 2017.

  • I was a member of PKU amateur athletic team, with the major of 5,000-meter race. I also enjoy 4*400 relay races. I won three gold medals at PKU athletic meetings, as well as one gold medal on 4*400 relay race with my great teammates at Track and Field Meeting of Beijing Universade.

  • I love poetry for its irresistible beauty! I often translates poems between Chinese and English (coming soon!).

  • I love Ludwig Wittgenstein and his theories.

  • I reviewed a book (Chapters of basic algorithm, data structures and mathematics) on Olympiad of Informatics, of which the author was my boyfriend Yudong (Victor) Li (a famous competitor Olympiad in Informatics and ACM-ICPC in China, known as lydrainbowcat). Click here to buy the book (in Chinese). We are also planning to translate it into English!

© 2018 Freda (Haoyue) Shi