Murong Yue | George Mason University

About Me

Hi, my name is Murong Yue (岳牧荣). I am a Ph.D. student at George Mason University (GMU), where I am luckily working with Prof. Ziyu Yao. I earned a master’s degree from the University of Southern California (USC) and a bachelor’s degree from Beijing Jiaotong University (BJTU). Before starting my Ph.D., I also worked as a Machine Learning Engineer at Alibaba AI Labs.

My research interests include:

LLM Reasoning: I am interested in how to dynamically adapt reasoning strategies based on the nature of the input problem. This includes selecting the appropriate LLM model, choosing the reasonable reasoning paradigm, and utilizing tools effectively.

Building and Evaluating LLM Agents for Interdisciplinary Applications: I am also passionate about how to construct and evaluate LLM-based agents for emerging tasks in interdisciplinary applications. Besdies, I am also working on the evaluation in open-ended tasks.

Trustworthy LLMs: I am also interested in uncertainty estimation and LLM jailbreaking.

News

[Apr. 2025] Check out our new preprint about the LLM batch prompting attack and the LLM persona simulation.
[Feb. 2025] One paper is accepted to AAAI 2025 AI4EDU workshop.
[Jan. 2025] One paper is accepted to ICLR 2025.
[Apr. 2024] I am awarded the Outstanding PhD Student Award by GMU, CS department.
[Jan. 2024] One paper is accepted to ICLR 2024.
[Dec. 2023] One paper is accepted to ICASSP 2024.
[Oct. 2023] Our paper about an open-source platform for ALM is accepted to EMNLP 2023 demo.
[Apr. 2023] GMU will organize MASC-SLL 2023, an annual NLP event in the Mid-Atlantic area! I will be serving as a Colloquium Organizer.

Publications

Preprint Can LLMs Simulate Personas with Reversed Performance? A Benchmark for Counterfactual Instruction Following

Sai Adith Senthil Kumar, Hao Yan, Saipavan Perepa,Murong Yue, Ziyu Yao

Preprint Efficient but Vulnerable: Benchmarking and Defending LLM Batch Prompting Attack

Murong Yue, Ziyu Yao

ICLR 2025 DOTS: Learning to Reason Dynamically in LLMs via Optimal Reasoning Trajectories Search

Murong Yue, Wenlin Yao, Haitao Mi, Dian Yu, Ziyu Yao, Dong Yu

AAAI Workshop 2025 MathVC: An LLM-Simulated Multi-Character Virtual Classroom for Mathematics Education

Murong Yue, Wenhan Lyu, Wijdane Mifdal, Jennifer Suh, Yixuan Zhang, Ziyu Yao

ICLR 2024 Large Language Model Cascades with Mixture of Thoughts Representations for Cost-efficient Reasoning

Murong Yue, Jie Zhao, Min Zhang, Liang Du, Ziyu Yao

ICASSP 2024 Cam LLM Find The Green Circle? Investigation And Human-guided Tool Manipulation For Compositional Generalization

Min Zhang, Jianfeng He, Shuo Lei, Murong Yue, Linhan Wang, Chang-Tien Lu

EMNLP 2023 Gentopia: A Collaborative Platform for Tool-Augmented LLMs

Binfeng Xu, Xukun Liu, Hua Shen, Zeyu Han, Yuhan Li, Murong Yue, Zhiyuan Peng, Yuchen Liu, Ziyu Yao, Dongkuan Xu

Award

Outstanding PhD Students, GMU CS Department

Recognition for Service on Hangs-On Generative AI Tutorial, GMU CEC

Services

Organizing Committee

MASC-SLL 2023 - 10th Mid-Atlantic Student Colloquium on Speech, Language and Learning

Reviewer

NeurIPS’23, NeurIPS’24, EMNLP'24, ICLR'25, AISTATS'25

Talk

[Feb. 2024] Title: LLM agent: build your own personal assistant, Invited talk at George Mason University Event for Hand-on Generative AI. Host: Amadra Shehu.

[Oct. 2023] Title: Large Language Model Cascades with Mixture of Thoughts Representations for Cost-efficient Reasoning, Invited talk at Microsoft Semantic Machines. Host: Hao Fang.