CV

Education

• B.S. in School of Computer Science, Fudan University, 2020-2024

Research Interests

• AI Safety&Alignment: Developing artificial general intelligence (AGI) to be more helpful, honest, and harmless, as well as more reliable, robust, and generally aligned with human values and follow human intent.

• Trustworthy ML/NLP: Unbiasedness, Robustness, Interpretability (How can we make sure that the model doesn’t confound by the potential bias existing in the training data and outputs conform to facts? Does the model has the ability to generalize well when cross-domain? …)

• Language Grounding in Multimodal Embodied Intelligence: Exploring how language grounding can be integrated with vision, speech, robotics, and other modalities to create embodied intelligent systems.

• Causality for NLP: Towards the well combination of Causality and NLP model to improve the efficiency of current statistical model with the capacity of casual inference.

Research Experience

• 2021.07-present: Fudan NLP Lab
  • Project 1: Causality & Implict Sentiment Analysis (published in COLING 2022)
  • Project 2: Causality & Domain generalization (published in LREC-COLING 2024)
  • Supervisor: Postdoc Jie Zhou and Prof. Xuanjing Huang

Publications

Causal Intervention Improves Implicit Sentiment Analysis

Siyin Wang, Jie Zhou, Changzhi Sun, Junjie Ye, Gui Tao, Qi Zhang and Xuanjing Huang

Proceedings of the 29th International Conference on Computational Linguistics (COLING 2022)

Domain Generalization via Causal Adjustment for Cross-Domain Sentiment Analysis

Siyin Wang, Jie Zhou, Qin Chen, Qi Zhang, Tao Gui, and Xuanjing Huang

The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)

Honors & Awards

• Contemporary Undergraduate Mathematical Contest in Modeling (CUMCM), First prize (Top 0.6% of 49242 teams), 2022.09

Service & Leadership

• 2022.02 - 2023.02: Deputy Director of Youth Volunteers’ Centre
  • School of Computer Science, Fudan University
  • Jishu Talk: co-founded the first technological talk by the youth, for the youth in Fudan University, which serves as a great platform for undergraduates to communicate and learn some basic skills not taught in class (like Python, Linux) as well as new fields (like ACM, CTF). 300+ participated in first (2022 Autumn).
• 2020.09 - present: Member of Fudan Poetry Society
  • Write poetry, critique the poetry of others, and join the 230 Anonymous Critic Association.