Shanyong Wang 王善泳

Hi there 👋 I'm Shanyong Wang, a senior undergraduate at ShanghaiTech University and an exchange student at UIUC (2024-2025), majoring in Computer Science.

I had a wonderful summer at Rutgers University working with Prof. Yongfeng Zhang. Previously, I was fortunate to be a member of the BLENDER Lab at UIUC, advised by Prof. Heng Ji and mentored by Xiusi Chen. I also closely collaborate with Prof. Ze Xiong at ShanghaiTech University.

My pronouns are He/Him/His.

News
2025
Our paper DecisionFlow has been accepted as EMNLP 2025 Findings. See you at Suzhou, China.
Aug 21
2024
Begin my exchange at UIUC.
Aug 21
Education
  • ShanghaiTech University
    Sept. 2022 - Jun. 2026
    Bachelor of Computer Science
  • University of Illinois Urbana-Champaign(UIUC)
    Aug. 2024 - May. 2025
    Visiting Student
    Computer Sciences
Experience
logo
Rutgers University
Advised by Prof. Yongfeng Zhang.
Jun. 2025 - Sept. 2025
Enhanced steerability in LLM preference alignment.
logo
University of Illinois Urbana-Champaign(UIUC)
Advised by Prof. Heng Ji and mentor Xiusi Chen.
Oct. 2024 - May. 2025
Designed a novel decision-modeling framework.
logo
ShanghaiTech Univeristy
Advised by Prof. Ze Xiong.
2024 - Present
Designed, modeled, and coded various virtual environments.
Publications
DecisionFlow: Advancing Large Language Model as Principled Decision Maker
DecisionFlow: Advancing Large Language Model as Principled Decision Maker

Xiusi Chen*, Shanyong Wang*, Cheng Qian*, Hongru Wang*, Peixuan Han, Heng Ji (* equal contribution)

EMNLP 2025 Findings

We propose DecisionFlow, a novel decision modeling framework that guides models to reason over structured representations of actions, attributes, and constraints. Rather than predicting answers directly from prompts, DecisionFlow builds a semantically grounded decision space and infers a latent utility function to evaluate trade-offs in a transparent, utility-driven manner. This process produces decisions tightly coupled with interpretable rationales reflecting the model's reasoning.

DecisionFlow: Advancing Large Language Model as Principled Decision Maker

Xiusi Chen*, Shanyong Wang*, Cheng Qian*, Hongru Wang*, Peixuan Han, Heng Ji (* equal contribution)

EMNLP 2025 Findings

We propose DecisionFlow, a novel decision modeling framework that guides models to reason over structured representations of actions, attributes, and constraints. Rather than predicting answers directly from prompts, DecisionFlow builds a semantically grounded decision space and infers a latent utility function to evaluate trade-offs in a transparent, utility-driven manner. This process produces decisions tightly coupled with interpretable rationales reflecting the model's reasoning.

About Me

I was born and raised in Nanjing, Jiangsu, China.