Yuxuan Li 李宇轩

I'm a second-year PhD student in the Human-Computer Interaction Institute (HCII) within Carnegie Mellon University (CMU)'s School of Computer Science, advised by Hirokazu Shirado and Sauvik Das.

Humans are increasingly delegating agency to machines such as language models in social contexts, giving rise to social agents. I lead research advancing the study of social agents through three main directions:

  • Deploying social agents at scale for useful social simulations that inform real-world practice.
    arXiv'25a
  • Using social agents as proxies to examine machine behavior.
    EMNLP'25 Oral, FAccT'25, arXiv'25b
  • Understanding how social agents influence human behavior and how they can be used to shape it.

I hold a BS in Computer Science from Tsinghua University, advised by Chun Yu and Yuanchun Shi. I was also a research intern at UC Berkeley, advised by Coye Cheshire.

Curriculum Vitae  /  Email  /  Google Scholar  /  LinkedIn  /  X  /  Bluesky

profile photo

Selected Papers

What Makes LLM Agent Simulations Useful for Policy? Insights From an Iterative Design Engagement in Emergency Preparedness thumbnail

What Makes LLM Agent Simulations Useful for Policy? Insights From an Iterative Design Engagement in Emergency Preparedness


Yuxuan Li, Sauvik Das, Hirokazu Shirado
Under review at CHI 2026
paper / cite

LLM agent simulations can be genuinely useful for policy. We work closely with policymakers over 16 months to design and build a 13,000-agent simulation system that directly informed and improved policy at CMU.
Spontaneous Giving and Calculated Greed in Language Models thumbnail

Spontaneous Giving and Calculated Greed in Language Models


Yuxuan Li, Hirokazu Shirado
EMNLP 2025 Main | The 2025 Conference on Empirical Methods in Natural Language Processing
Oral Presentation | Extensive Media Coverage
paper / video / selected media coverage / cite

Reasoning models are greedier. We find that reasoning models consistently exhibit lower cooperation and reduced norm-enforced punishment, mirroring human tendencies of "spontaneous giving and calculated greed".
Actions Speak Louder than Words: Agent Decisions Reveal Implicit Biases in Language Models thumbnail

Actions Speak Louder than Words: Agent Decisions Reveal Implicit Biases in Language Models


Yuxuan Li, Hirokazu Shirado, Sauvik Das
FAccT 2025 | ACM Conference on Fairness, Accountability, and Transparency
paper / video / cite

LLMs are increasingly implicitly biased. We find that when simulating human behavior (actions), more advanced models exhibit stronger sociodemographic biases, even though these biases appear reduced when measured explicitly through Q&A (words).
HiddenBench: Assessing Collective Reasoning in Multi-Agent LLMs via Hidden Profile Tasks thumbnail

HiddenBench: Assessing Collective Reasoning in Multi-Agent LLMs via Hidden Profile Tasks


Yuxuan Li, Aoi Naito, Hirokazu Shirado
Under review at ICLR 2026
paper / cite

Multi-agent LLM systems fail at collective reasoning, much like human groups. We demonstrate this by formalizing the Hidden Profile paradigm from social psychology and constructing a scalable 65-task benchmark based on this formalization.
A Human-Computer Collaborative Tool for Training a Single Large Language Model Agent into a Network through Few Examples thumbnail

A Human-Computer Collaborative Tool for Training a Single Large Language Model Agent into a Network through Few Examples


Lihang Pan*, Yuxuan Li*, Chun Yu, Yuanchun Shi (*denotes equal contribution)
arXiv 2024
paper / cite

We make it easy for novices to build powerful multi-agent LLM systems (MAS). We introduce EasyLAN, a human–AI collaborative system that helps users construct MAS using only a few examples.
Say Your Reason: Extract Contextual Rules In Situ for Context-aware Service Recommendation thumbnail

Say Your Reason: Extract Contextual Rules In Situ for Context-aware Service Recommendation


Yuxuan Li, Jiahui Li, Lihang Pan, Chun Yu, Yuanchun Shi
arXiv 2024
paper / cite

We make it easy to control service recommendations on users’ mobile phones in situ. We introduce SayRea, an interactive system that helps extract contextual rules for personalized, context-aware service recommendations in mobile scenarios using LLMs.