I am a postdoctoral scholar at Data Science Institute, University of Chicago, working with Prof. Chenhao Tan and Prof. James Evans. I obtained my Ph.D. degree in Wangxuan Institute of Computer Technology, Peking University, advised by Prof. Yansong Feng in 2025. I was a visiting Ph.D. student at University of California, Los Angeles, working with Prof. Kai-Wei Chang in 2023. I received the BSc degree from School of Electronic Engineering and Computer Science, Peking University and BEc degree from National School of Development in 2020.

[Google Scholar] [Github] [X] [CV]

My research aims to enable trustworthy decision-making and rigorous scientific discovery through causally intelligent AI. I develop methods that strengthen causal reasoning in large language models, helping them understand underlying mechanisms, draw conclusions supported by evidence, and recognize when that evidence is insufficient. Building on these capabilities, I explore how AI can advance scientific discovery by integrating existing knowledge, generating new hypotheses, and testing them through new experiments.

Towards a Universal Causal Reasoner
NeurIPS 2026 (Spotlight)
Qirun Dai*, Xiao Liu*, Jiawei Zhang*, Dylan Zhang, Hao Peng, Chenhao Tan
Causal Reasoning, Trustworthy AI
[paper] [data]

Introspection Tools Help LLMs Understand and Control Themselves
NeurIPS 2026
Xiao Liu*, Junsol Kim*, Shiyang Lai*, Jacy Reese Anthis*, Siyang Wu, Chenhao Tan, James Evans
Trustworthy AI

RefTool: Enhancing Model Reasoning with Reference-Guided Tool Creation
ICLR 2026
Xiao Liu, Da Yin, Zirui Wu, Yansong Feng
Causal Reasoning
[paper] [code]

Heuristic-Based Ideation for Guiding LLMs Toward Structured Creativity
ICLR 2026 Blogpost
Xiao Liu*, Haokun Liu*, Chenhao Tan
AI for Scientific Discovery
[blog]

From Single to Societal: Analyzing Persona-Induced Bias in Multi-Agent Interactions
AAAI 2026 (Oral)
Jiayi Li*, Xiao Liu*, Yansong Feng
Trustworthy AI
[paper] [code]

Improving Research Idea Generation Through Data: An Empirical Investigation in Social Science
AI4Science Workshop at Neurips 2025 (Spotlight)
Xiao Liu, Xinyi Dong, Xinyang Gao, Yansong Feng, Xun Pang
AI for Scientific Discovery
[paper]

Eliciting and Improving the Causal Reasoning Abilities of Large Language Models with Conditional Statements
Computational Linguistics 2025, Presented at ACL 2025
Xiao Liu, Da Yin, Chen Zhang, Yansong Feng, Dongyan Zhao
Causal Reasoning
[paper]

Are LLMs Capable of Data-based Statistical and Causal Reasoning? Benchmarking Advanced Quantitative Reasoning with Data
ACL 2024 (Findings), AI4Research workshop @ IJCAI 2024 Highlight Paper
Xiao Liu, Zirui Wu, Xueqing Wu, Pan Lu, Kai-Wei Chang, Yansong Feng
Causal Reasoning
[paper] [code] [website]

CASA: Causality-driven Argument Sufficiency Assessment
NAACL 2024
Xiao Liu, Yansong Feng, Kai-Wei Chang
Causal Reasoning, Trustworthy AI
[paper] [code]

Dynosaur: A Dynamic Growth Paradigm for Instruction-Tuning Data Curation
EMNLP 2023
Da Yin*, Xiao Liu*, Fan Yin*, Ming Zhong*, Hritik Bansal, Jiawei Han, Kai-Wei Chang
[paper] [code]

The Magic of IF: Investigating Causal Reasoning Abilities in Large Language Models of Code
ACL 2023 (Findings)
Xiao Liu, Da Yin, Chen Zhang, Yansong Feng, Dongyan Zhao
Causal Reasoning
[paper] [code]

Counterfactual Recipe Generation: Exploring Compositional Generalization in a Realistic Scenario
EMNLP 2022
Xiao Liu, Yansong Feng, Jizhi Tang, Chengang Hu, Dongyan Zhao
Causal Reasoning
[paper] [code]

Things not Written in Text: Exploring Spatial Commonsense from Visual Signals
ACL 2022
Xiao Liu, Da Yin, Yansong Feng, Dongyan Zhao
[paper] [code]

Everything Has a Cause: Leveraging Causal Inference in Legal Text Analysis
NAACL 2021 (Oral)
Xiao Liu*, Da Yin*, Yansong Feng, Yuting Wu, Dongyan Zhao
Causal Reasoning, Trustworthy AI
[paper] [code]

Building an Atlas of Social Experiments to Link Studies, Reconcile Conflicts, and Bridge Gaps
Jiawei Zhang*, Honglin Bao*, Pengda Wang*, Alex Yan*, Xiao Liu^✉, James Evans
AI for Scientific Discovery
[paper]

Markets, Not Planners: Decentralized Orchestration of LLM Agents with Private Information
Xiao Liu*, Haoyang Li*, Songwei Li, Hongbo Fang, Fengli Xu, Feng Shi, James Evans
Trustworthy AI
[paper]

Contemporary AI lacks the imagination to diverge or negate in science
Honglin Bao, Siyang Wu, Xiao Liu, Sida Li, Shiyun Cao, James Evans
AI for Scientific Discovery
[paper]

CausaLab: A Scalable Environment for Interactive Causal Discovery Toward AI Scientists
Junlin Yang, Dylan Zhang, Xiangchen Song, Qirun Dai, Xiao Liu, Yuen Chen, Aniket Vashishtha, Jing Shi, Chenhao Tan, Hao Peng
Causal Reasoning, AI for Scientific Discovery
[paper]

• Area Chair ACL Rolling Review 2024-2026
• Senior Program Committee AAAI 2026
• Reviewer ICLR, NeurIPS, ACL, EMNLP, NAACL, COLM, CSCW, Computational Linguistics, TKDD, TKDE, IPM
• Volunteer ACL, EMNLP, NAACL
• Co-Organizer AI & Scientific Discovery Online Seminar

• Mentor Data Science Clinic, University of Chicago, Winter & Spring 2026
• Workshop Instructor and Textbook Co-Author Data Thinking and Analytical Methods for the Social Sciences, Peking University, Spring 2025
• Teaching Assistant Foundations of Natural Language Processing, Peking University, Spring 2024
• Teaching Assistant Empirical Methods for Natural Language Processing, Peking University, Spring 2023
• Teaching Assistant Introduction to Computation, Peking University, Fall 2020

• Rising Star in Data Science, 2026
• DSI Faraco Family Postdoctoral Fellowship, University of Chicago, 2026
• Outstanding Graduate, Peking University, 2025
• Tung Scholarship, Peking University, 2022
• National Scholarship, Peking University, 2018

Website design: ✩ ✩ ✩