Xiao Liu (刘潇)
Email: lxlisa [AT] pku [DOT] edu [DOT] cn

I am a fifth-year Ph.D. student at Wangxuan Institute of Computer Technology, Peking University, advised by Prof. Yansong Feng and Prof. Dongyan Zhao. I was fortunate to visit University of California, Los Angeles and collaborate with Prof. Kai-Wei Chang and Prof. Nanyun Peng in 2023. I received my 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] [Twitter]

  Research Interests

I am deeply engaged in advancing the reasoning abilities of language models, aiming to unravel real-world challenges. Particularly, my research focuses on the capabilities of causal reasoning and compositional generalization.

  • Causal reasoning: explore LLMs' abilities to conduct data-based causal reasoning and commonsense causal reasoning (ACL 2024, ACL 2023), improve the performance on logical and legal reasoning tasks with causal inference (NAACL 2024, NAACL 2021).
  • Compositional generalization: evaluate and enhance compositional generalization abilities of LLMs on real-world procedural text (EMNLP 2023, EMNLP 2022).
I am also interested in synthetic data generation (EMNLP 2023), low-resource languages (ACL 2024), commonsense reasoning (ACL 2022), and knowledge representation (ACL 2020, TETCI 2020, EMNLP 2019, IJCAI 2019).
  Publications

QUDSELECT: Selective Decoding for Questions Under Discussion Parsing
EMNLP 2024
Ashima Suvarna*, Xiao Liu*, Tanmay Parekh, Kai-Wei Chang, Nanyun Peng
[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
[paper] [code] [website]

Teaching Large Language Models an Unseen Language on the Fly
ACL 2024 (Findings)
Chen Zhang, Xiao Liu, Jiuheng Lin, Yansong Feng
[paper] [code] [website]

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

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
(* Equal Contribution)
[paper] [code] [website]

DiNeR: A Large Realistic Dataset for Evaluating Compositional Generalization
EMNLP 2023
Chengang Hu, Xiao Liu, Yansong Feng
[paper] [code] [model]

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
[paper] [code]

How Many Answers Should I Give? An Empirical Study of Multi-Answer Reading Comprehension
ACL 2023 (Findings)
Chen Zhang, Jiuheng Lin, Xiao Liu, Yuxuan Lai, Yansong Feng, Dongyan Zhao
[paper] [code]

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

Dual-Channel Evidence Fusion for Fact Verification over Texts and Tables
NAACL 2022
Nan Hu, Zirui Wu, Yuxuan Lai, Xiao Liu, Yansong Feng
[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
Xiao Liu*, Da Yin*, Yansong Feng, Yuting Wu, Dongyan Zhao
[paper] [code]

A Graph Learning Based Approach for Identity Inference in Dapp Platform Blockchain
IEEE Transactions on Emerging Topics in Computing 2020
Xiao Liu, Zaiyang Tang, Peng Li, Song Guo, Xuepeng Fan, Jinbo Zhang
[paper]

Neighborhood Matching Network for Entity Alignment
ACL 2020
Yuting Wu, Xiao Liu, Yansong Feng, Zheng Wang, Dongyan Zhao
[paper] [code]

Interactive Multi-grained Joint Model for Targeted Sentiment Analysis
CIKM 2019
Da Yin*, Xiao Liu*, Xiaojun Wan
[paper]

Jointly Learning Entity and Relation Representations for Entity Alignment
EMNLP 2019
Yuting Wu, Xiao Liu, Yansong Feng, Zheng Wang, Dongyan Zhao
[paper] [code]

Relation-aware Entity Alignment for Heterogeneous Knowledge Graphs
IJCAI 2019
Yuting Wu, Xiao Liu, Yansong Feng, Zheng Wang, Rui Yan, Dongyan Zhao
[paper] [code]

  Academic Services
  • Area Chair ACL Rolling Review 2024
  • Reviewer ACL Rolling Review 2021-2024, COLM 2024, ACL 2020-2023, EMNLP 2021-2022, NAACL 2021, Computational Linguistics, TKDE, IPM
  • Volunteer ACL 2024, NAACL 2024, EMNLP 2022, ACL 2021, NAACL 2021, EMNLP 2020
  Teaching
  • 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 B, Peking University, Fall 2020
  Selected Awards
  • Tung Scholarship, Peking University, 2022
  • Merit Student of Peking University, 2022, 2017-2019
  • Wangxuan Scholarship, Wangxuan Institute of Computer Technology, 2020
  • National Scholarship, Peking University, 2018
  Contact

Wangxuan Institute of Computer Technology, Peking University
No. 128 Zhongguancun North Street
Haidian District, Beijing, 100871
lxlisa [at] pku.edu.cn


Website design: