About

I am an incoming CS PhD student at Johns Hopkins University, advised by Prof. Tianmin Shu. I obtained my B.A. in Honors CS and Mathematics, and received CS Prize for the Most Promising Student from NYU. I was previously a visiting student at MIT working with Prof. Josh Tenenbaum and Prof. Tianmin Shu.

Here are some research directions I’m excited about:

Deliberate Reasoning: combine language models, agent models, and world models in a cognitively grounded way, and promote more robust and versatile reasoning across multiple modalities.
AI Agents: build the next generation of language agents that can interact with the world through reasoning, planning, managing memory, and learning from human and environment feedback.
Social Intelligence: create agents that can perceive, reason about, learn from, and respond to affect, behavior, and cognition of humans or other agents, and foster strategic collaborations among agents.

Besides academics, I enjoy roaming in movies and books. I am an avid reader of history and scientific fiction, as well as a dedicated writer read by more than 16,000,000 times on different platforms. I also enjoy playing “Go”, piano, badminton, hiking and so much more.

News

[May, 2024] MMToM-QA is accepted to ACL 2024 with four top ratings. See you in Bangkok!
[May, 2024] Released our position paper for ICLR 2024 AGI Workshop.
[Dec, 2023] I am attending NeurIPS 2023. Feel free to reach out if you’d like to have a chat!
[May, 2023] I am joining MIT Cocosci and CSAIL as a visiting student
[May, 2023] Released – SOTA diffusion models with 2x speedup
[Apr, 2023] Received NYU Computer Science Prize for the Most Promising Student (1 person/year)

Research

MMToM-QA: Multimodal Theory of Mind Question Answering
Chuanyang Jin, Yutong Wu, Jing Cao, Jiannan Xiang, Yen-Ling Kuo, Zhiting Hu, Tomer Ullman, Antonio Torralba, Joshua Tenenbaum, Tianmin Shu
ACL 2024
Nominated for Outstanding Paper Award
paper / project / code / slides / twitter

How Far Are We From AGI?
Chuanyang Jin*, Tao Feng*, Jingyu Liu*, Kunlun Zhu*, Haoqin Tu, Zirui Cheng, Guanyu Lin, Jiaxuan You
Under Review
Position Paper for ICLR 2024 AGI Workshop
paper / project / workshop / twitter

Neural Amortized Inference for Nested Multi-agent Reasoning
Kunal Jha, Tuan Anh Le, Chuanyang Jin, Yen-Ling Kuo, Joshua Tenenbaum, Tianmin Shu
AAAI 2024
Oral Presentation at AAAI Summer Symposium
paper / project / code / slides

Dynamics of RNA Localization to Nuclear Speckles are Connected to Splicing Efficiency
Jinjun Wu*, Yu Xiao*, Yunzheng Liu*, Li Wen, Chuanyang Jin, Shun Liu, Sneha Paul, Chuan He, Oded Regev, Jingyi Fei
Under Review
paper

Projects

Fast Diffusion Transformers
DiT models with 2x speedup and 60% less memory, 500+ GitHub Stars.
project

The Cultural Psychology of Large Language Models: Is ChatGPT a Holistic or Analytic Thinker?
Chuanyang Jin, Songyang Zhang, Tianmin Shu, Zhihan Cui
paper