Sijie Cheng

Sijie Cheng

Ph.D. Candidate

Tsinghua University

Biography

Sijie Cheng is a Ph.D. candidate at the School of Computer Science and Technology, Tsinghua University, Beijing, China. She is advised by Prof. Yang Liu at THUNLP and Institute for AI Industry Research (AIR). She received her Master’s degree from Fudan University in 2023, advised by Prof. Yanghua Xiao at Knowledge Work Lab. Sijie’s previous work focused on the analysis and exploitation of tacit knowledge within foundation models. Currently, she devotes herself to working on grounding foundation models in the physical world. She welcomes collaboration opportunities and encourages interested individuals to reach out to her.

Download my resumé .

Interests
  • World Model
  • Foundation Models
  • Embodied Artificial Intelligence
Education
  • Ph.D. in Computer Science and Technology, 2023

    Tsinghua University

  • M.Sc in Software Engineering, 2020

    Fudan University

  • B.Sc. in Software Engineering, 2016

    Chongqing University of Posts and Telecommunications

News

Experience

 
 
 
 
 
Pre-train and Multi-modal Group, 01.AI
Research Intern
August 2023 – Present Beijing, China
Supervised Fine-tuning and Video-Language Models.
 
 
 
 
 
Investment Department, Sinovation Ventures
Investment Intern
February 2023 – Present Beijing, China
Investigation for Artificial Intelligence, working with Bobing Ren.
 
 
 
 
 
Natural Language Processing Group, Pujiang Lab
Research Intern
March 2022 – December 2022 Shanghai, China
Free-text explanation with LLMs, working with Prof. Lingpeng Kong.
 
 
 
 
 
Institue for AI Industry Research, Tsinghua University
Visiting Fellow
July 2021 – February 2022 Beijing, China
LLMs as Continual Knowledge Bases, advised by Prof. Yang Liu and Prof. Yang Liu.
 
 
 
 
 
Natural Language Understanding Group, Meituan NLP Center
Research Intern
November 2020 – June 2021 Beijing, China
Taxonomy Extraction by LLMs, working with Rui Xie.
 
 
 
 
 
Text Intelligence Lab, Westlake University
Visiting Fellow
September 2019 – September 2020 Hangzhou, China
Analysis of commonsense knowledge in LLMs, advised by Prof. Yue Zhang.

Recent Publications

Quickly discover relevant content by filtering publications.
(2023). Modeling Adversarial Attack on Pre-trained Language Models as Sequential Decision Making. In ACL 2023.

PDF Cite

(2023). Prompt-Guided Retrieval Augmentation for Non-Knowledge-Intensive Tasks. In ACL 2023.

PDF Cite

(2023). Say What You Mean! Large Language Models Speak Too Positively about Negative Commonsense Knowledge. In ACL 2023.

PDF Cite Code

(2022). Can Pre-trained Language Models Interpret Similes as Smart as Human?. In ACL 2022.

PDF Cite Code

(2022). Unsupervised Editing for Counterfactual Stories. In AAAI 2022.

PDF Cite Code

Projects

OpenChat
Advancing Open-source Language Models with Imperfect Data
OpenChat

Awards

Excellent Graduates
National Scholarship
1st Place in Women’s Basketball Graduate School Cup