Ruohan Wang

Hi there👋 I'm Ruohan (Demi) Wang, an undergraduate student at Tongji University.

In a short term, I'm focused on conducting research on large language models (LLMs) and learning to be a full-stack independent developer. In the long run, I hope to build a startup and contribute to the development of AGI.

I've gained research experience through internships at ByteDance and OSU NLP Group (supervised by Prof. Yu Su). My current interests lie in language agents and multimodal models. I'm also interested in participating in AI-related hackathons.

News
2025
UGround was accepted as ICLR 2025 Oral (1.8%)!
Feb 11
I'm on China's top 1 intellectual game show, The Brain Season 12! Check it out on Youtube🙌
Jan 06
2024
Proud to share that I ranked 352nd out of 50,000+ in the Alibaba Global Mathematics Competition!
Jun 13
2023
We won 2nd place at the Baidu AGI HACKATHON!
Dec 13
I'm starting my gap year and joining ByteDance E-Commerce team as a Machine Learning Intern!
Oct 20
Started my visit at UC Berkeley. Go Bears!
Jan 06
2022
I received Berkeley Global Access Scholarship!
Nov 20
Education
  • Tongji University
    Sept. 2020 - Jun. 2025
    Bachelor of Software Engineering
  • University of California, Berkeley
    Jan. 2023 - Aug. 2023
    Visiting Student
    Electrical Engineering and Computer Sciences
Experience
logo
The Ohio State University
Research Intern
Apr. 2024 - Oct. 2024
Improving the performance of
GUI Agents across platforms
logo
ByteDance
Machine Learning Engineer Intern
Oct. 2023 - Feb. 2024
Enhancing e-commerce violation detection with LLMs
logo
University of California, Berkeley
Undergraduate Researcher
Feb. 2023 - May 2023
Building and analyzing Irish
drama corpus
Publications
Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI Agents
Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI Agents

Boyu Gou, Ruohan Wang, Boyuan Zheng, Yanan Xie, Cheng Chang, Yiheng Shu, Huan Sun, Yu Su

ICLR 2025 Oral (1.8%)

We present SeeAct-V, a human-like, vision-only GUI agent framework, and UGround, a SOTA GUI visual grounding model trained on cost-effective synthetic data, marking the first practical demonstration of SOTA performance for vision-only GUI agents.

Paper Code Homepage Twitter
Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI Agents

Boyu Gou, Ruohan Wang, Boyuan Zheng, Yanan Xie, Cheng Chang, Yiheng Shu, Huan Sun, Yu Su

ICLR 2025 Oral (1.8%)

We present SeeAct-V, a human-like, vision-only GUI agent framework, and UGround, a SOTA GUI visual grounding model trained on cost-effective synthetic data, marking the first practical demonstration of SOTA performance for vision-only GUI agents.

Paper Code Homepage Twitter
Selected Projects
Life Buddy

Life Buddy

2nd Winner @Baidu AGI Hackathon

Your personal lifestyle assistant for a budget-friendly life. Discover affordable dining, nail salons, entertainment, and more—all tailored to your preferences.

Official Intro Post
Let them cook

Let them cook

@CalHack 9.0

Nearly 40% of food in the U.S. is wasted. Reduce yours by Leting Them Cook, a recipe AI tool that generates fridge-cleaning recipes that are both beneficial to your wallet and the world.

Devpost Code
cowbell.ai

cowbell.ai

@Outside LLMs Hackathon

Unopinionated music production assistant. Any instrument, anywhere.

Devpost Code Demo
Healper

Healper

An AI-powered platform for psychological counseling support—assess your mental state, connect with professional counselors, and streamline record-keeping.

Frontend Backend
Honors & Awards
No. 352 (Top 1%), Alibaba Global Mathematics Competition Qualifying Round
Jun. 2024
First Prize in International Exchange Scholarship, Tongji University
Mar. 2024
Berkeley Global Access Scholarship, Berkeley Extension
Nov. 2022
Scholarship for Outstanding Students, Tongji University
Nov. 2022 & Oct. 2021
Second Prize, China National High School Mathematics Competition
Sep. 2019
Honor Roll (Top 5%), American Mathematics Competition 12
Feb. 2018
About Me

I'm INTP, Slytherin.

I played piano 🎹 for 10 years, and now I'm a hip-hop songwriter 🎵 —listen to my music here.

I enjoy sharing my experiences on social media 📱, so far I've got over 50k likes and saves⭐️.

I wanted to be a writer when I was young, and published a short story 📖.

In my free time, I dance 💃, snowboard 🏂, travel ✈️, and make cocktails 🍸.

I love exploring new stuffs 🌈. I've taken solo trip to over 10 places 🗺️, and have tried over 10 different hair colors.