Guangyuan Jiang

Yuanpei College, Peking University


I am Guangyuan Jiang (姜广源), a junior undergrad (‘25) majoring in Artificial Intelligence at Yuanpei College, Peking University.

I’m currently a visiting student at MIT Computational Cognitive Science Lab and Computational Psycholinguistics Lab. I’m fortunate to be advised by Prof. Josh Tenenbaum and Prof. Roger Levy.

At Peking University, I am a member of the Tong Class (an honorary pilot class in AI). I’m also affiliated with the PKU Cognitive Reasoning Lab and am grateful to work with Prof. Yixin Zhu.

My research interests encompass a range of topics, including abstraction, language and concept learning, abstract reasoning, and machine behavioral study. My overarching objective is to contribute to the development of machines that can learn and reason in a manner akin to human cognition. Currently, I focus on exploring abstraction in language evolution and cultural transmission, how language guides abstraction learning, and the acquisition of words and concepts.

Selected Publications

  1. CogSci 2024 Talk
    Finding structure in logographic writing with library learning
    Guangyuan Jiang, Matthias Hofer, Jiayuan Mao, Lionel Wong, Josh Tenenbaum, and Roger Levy
    In CogSci, 2024
  2. ICML 2023
    MEWL: Few-shot multimodal word learning with referential uncertainty
    Guangyuan Jiang, Manjie Xu, Shiji Xin, Wei Liang, Yujia Peng, Chi Zhang, and Yixin Zhu
    In ICML, 2023
  3. NeurIPS 2023 Spotlight
    Evaluating and Inducing Personality in Pre-trained Language Models
    Guangyuan Jiang*, Manjie Xu*, Song-Chun Zhu, Wenjuan Han, Chi Zhang, and Yixin Zhu
    In NeurIPS, 2023
  4. NeurIPS 2023 D&B
    Interactive Visual Reasoning under Uncertainty
    Manjie Xu*, Guangyuan Jiang*, Wei Liang, Chi Zhang, and Yixin Zhu
    In NeurIPS Datasets and Benchmarks, 2023