I am now a research assistant in THUNLP Lab at Tsinghua University, and am currently looking for a PhD position working on computer vision.
I received my master’s degree in Computer Technology from Tsinghua University in 2024, advised by Prof.Zhi Wang . Previously I got my bachelor’s degree in Computer Science and Technology from Renmin University of China in 2021.
My past research experience has involved downstream works with GANs and Diffusion Models, including applying Deepfake in live streaming and procedural content generation in games. I am currently interested in Embodied AI, with a special focus on interactive scene generation. For more details, please see my CV.
🔥 News
- 2024.07: I got my master’s degree 🥳, and join THUNLP lab as a RA.
- 2024.03: One paper is accepted by ICME 2024.
- 2023.12: 🎉 My first work is accepted by AAAI 2024! Lucky me 🤪!
📝 Publications
EmbodiedEval: Evaluating Multimodal LLMs as Embodied Agents
Zhili Cheng*, Yuge Tu*, Ran Li*, Shiqi Dai*, Jinyi Hu*, Shengding Hu, Jiahao Li, Yang Shi, Tianyu Yu, Weize Chen, Lei Shi, Zhiyuan Liu, Maosong Sun
- We propose a benchmark for Multimodal LLMs as embodied agents with 328 distinct tasks within 125 varied 3D scenes.
- The tasks are organized into five categories: navigation, object interaction, social interaction, attribute question answering, and spatial question answering to assess different embodied capabilities of Multimodal LLMs.
Procedural Level Generation with Diffusion Models from a Single Example
Shiqi Dai, Xuanyu Zhu, Naiqi Li, Tao Dai, Zhi Wang
- We propose a diffusion-based generative model learning from a single example in Minecraft and Super Mario Bros.
- Our model suits a wide range of level structures with fewer artifacts than GAN-based approaches.
Collaborative Edge Caching in LEO Satellites Networks: A MAPPO Based Approach
Mingzhou Wu, Shiqi Dai, Han Hu, Zhi Wang
- We model the movement of satellites into a Partially Observable Markov Decision Process and design a multi-agent reinforcement learning algorithm specifically tailored for content caching.
📖 Educations
- 2021.09 - 2024.06, Master of Computer Technology, Tsinghua University, Beijing, China.
- 2017.09 - 2021.06, Bachelor of Computer Science and Technology, Renmin University of China, Beijing, China.
- 2014.09 - 2017.06, Guiyang No.1 High School, Guizhou, China.
👩🏻🏫 Teaching Assistants
- Distributed Machine Learning, 2024 Spring.
- Literature Searching and Paper Writing, 2023 Spring.
- Big Data System (B), 2022 Fall.
💻 Internships
- 2024.07 - Present, Research assistant in Emobodied AI, THUNLP Lab, Beijing, China.
- 2023.12 - 2024.03, AIGC Intern (remote), Tencent, Shenzhen, China.
🎖 Honors and Awards
- 2021.06, Outstanding Graduate of School of Information (Top 5%), Renmin University of China
- 2020.05, Meritorious Winner (Top 8%) in Mathematical Contest In Modeling (MCM/ICM)
- 2019.10, Second Prize in China Undergraduate Mathematical Contest in Modeling (CUMCM)
🍭 Hobbies
I enjoy performing. I reached the amateur level 10 in electronic keyboard🎹 when I was 13. I am currently teaching myself to play the ukulele🎸. Additionally, I play tennis🎾 and badminton🏸 with friends every week.