Hi~ I am Yusen Zhang, a Postdoctoral Research Scientist in Computer Science at Columbia University, advised by faculty in the DAPLab (including Zhou Yu, Eugene Wu, David Blei, Daniel Hsu, etc.). I obtained my PhD degree from Penn State University, advised by Dr. Rui Zhang. I have done industry research internships at Amazon, Microsoft, and Google. I also worked closely with Dr. Dragomir Radev. I received my master’s degree from Emory University, advised by Dr. Jinho D. Choi. I am fortunate to be the recipient of the SAC Award of NAACL 2025 and the Best Paper Award of a workshop at AAAI 2025.

I am also open to academic collaborations—please feel free to reach out if you’re interested in working together! Let’s build the future of LLM together!

Research

My research interests lie in AI, NLP, and LLMs. Specifically, I am interested in building distributed Multi-Agent LLM Systems for Work (MA-LLM4Work), enabling multi-agent systems to function as collaborative organizational entities, similar to human society, to automate or augment human work. It involves two parts of research in diverse domains:

Scalable, efficient, trustworthy, and collaborative Large Language Model Agent systems

From a technical perspective, I aim to explore how multiple models can work together intelligently (Agent × NLP/LLM). For instance:

• How can we design mechanisms for efficient, accurate, and context-aware communication between LLMs?
• How can we harness collective intelligence across multiple agents for complex reasoning and high-stakes decision-making?
• What technological, ethical, and policy frameworks are needed for the responsible integration of such systems into society?

Make the Agent and Agent Systems work in real-world organizations to automate humans’ work.

From an application perspective, I build agents as effective members of human organizations, collaborating with people and other agents. This requires advances across multiple intersections:

• Agent × Database: Designing reliable memory systems that enable agents to pursue long-term goals.
• Agent × ML Theory: Understanding why and when multi-agent systems outperform single agents.
• Agent × HCI: Creating smooth human–agent collaboration in practical workflows.
• Agent × CV/Robotics: Developing actionable agents that can use computers, perceive visual environments, and make autonomous decisions.
• Agent × Business: Modeling humans as digital-agent counterparts (digital twins) that can work on their behalf.

My ultimate goal is to build an Agent Society — a user-centric ecosystem where multiple AI entities collaborate, communicate, and assist both humans and other agents in complex, real-world environments.

News

[2025.10] We organize the WMAC 2026 workshop! AAAI 2026 Bridge Program on Advancing LLM-Based Multi-Agent Collaboration. Welcome to submit, discuss, and brainstorm!

[2025.10] I passed my Phd defense, thanks to everyone who supports and helps me in my PhD study! Next, it is very excited to join DAPLab to work on multi-agent systems for work!

[2025.7] PSU News reports three recent projects I particiated in! Thanks, PSU!

[2025.6] One paper on Geometry Understanding of VLM is accepted at COLM 2025. Thanks to Ryo and all the collaborators!

[2025.6] One paper on “Visual Long Context” of VLM is accepted at ICCV 2025. Thanks to all my collaborators!

[2025.5] Two papers were accepted at ACL 2025 Findings. Congrats Xiaoxin and Xi!

[2025.5] One paper on fair summarization was accepted at NACCL 2025, and won the SAC award (25/2346)! Thanks Haoyuan!

[2025.5] AAAR-1.0 was accepted at ICML 2025, and won Best Paper Award at the 2nd AI4Research workshop @ AAAI 2025. Thanks Renze!

[2025.1] Google Blog reports our Chain-of-Agent! Thanks, Google!

[2025.1] One paper was accepted at ICLR 2025, and one was accepted at NAACL 2025. Thanks to Sarkar and Haoyuan Li!

[2024.9] One paper on Multi-Agent LLMs is accepted at NeurIPS 2024. Thanks to all my collaborators!

[2024.7] One paper on LLM error detection is accepted at COLM 2024. Thanks to Ryo and all the collaborators!

[2024.3] Got one paper accepted at NAACL 2024. Thanks to all my collaborators!

[2023.8] Got one paper accepted at EMNLP 2023. Thanks Nan Zhang and other collaborators!

[2023.8] Start Student Researcher Internship at Cloud AI Team @ Google Research, advised by Dr. Ruoxi Sun!

[2023.5] Got one paper accepted at ACL 2023. Thanks to all my collaborators!

[2023.4] Oral talk at MASC-SLL 2023 at Georgetown Mason University (Arlington campus)!

[2023.2] Got one paper accepted at TACL 2023 (acceptance type (a) accepted as is, ~= 5%!). Thanks to all my collaborators!

[2022.12] Give a talk at Westlate NLP on Controllable Text Summarization. Thanks for having me!

[2022.7] Receive Dr. Tse-Yun Feng Graduate Student Award @ PSU CSE. Thanks everyone!

[2022.6] Start Applied Scientist Internship at AWS AI lab @ Amazon Web Service, advised by Dr. Xiaofei Ma!

[2022.2] Got two papers accepted at ACL 2022. Thanks to all my collaborators!

[2022.2] Start Research Internship at Cognitive Service Research Group @ Microsoft Research, advised by Dr. Yang Liu.

[2021.8] Got two papers accepted at EMNLP 2021. Thanks to all my collaborators!

[2021.8] Started my PhD study at NLP Lab @ Penn State. I am fortunate to be supervised by Prof. Rui Zhang!

[2021.5] Got one paper accepted at ACL 2021. Thanks to all my collaborators!

[2021.1] Started to work as a research assistant at Lily Lab @ Yale University, advised by Dr. Dragomir Radev.