Yuxuan Jiang

PhD StudentLLMs & Tool-Integrated ReasoningNLP, LLM Reasoning

prof_pic.jpg

Yuxuan Jiang

yuxuanj1@umbc.edu


Affiliations

University of Maryland, Baltimore County
Ph.D. in Computer Science
2023 - Present
New York University
MS. in Computer Science
2021-2022
Advisor: Dr. He He
Grade: 3.6/4.0
Beijing Language and Culture University
B.Sc. in Information Systems
Fall 2016 - Fall 2020

Professional Services

Yuxuan Jiang is a Computer Science PhD student under Dr. Francis Ferraro at the University of Maryland, Baltimore County (UMBC).
His research centers on tool-integrated reasoning, LLM Memorization, and efficient large language models.

His ongoing projects include building large-scale reasoning benchmarks and studying how reasoning efficiency relates to model memory and inference balance.

Previously, Yuxuan earned his M.S. in Computer Science from New York University and B.A. in Information Systems from Beijing Language and Culture University.
He also interned at Xiaomi, where he worked on large-scale dialogue systems and commonsense reasoning integration, and at the National Language Resource Monitoring Center, where he led semantic role labeling and input-correction research.

He has authored several papers on large language model reasoning and evaluation, including works on memorization mitigation and reasoning-step compression.

I am currently open to research collaborations in large language models reasoning. Please email me if you’d like to discuss a project.

Yuxuan has a strong background in deep learning, natural language processing, with experience in PyTorch, Hugging Face, TensorFlow, and cloud deployment (AWS / GCP).
He has also been recognized for his contributions to research on reasoning efficiency, skill-aware step pruning, and distilled reasoning models in recent ACL and EMNLP submissions.

Research Interests

        ✓ Tool-Integrated Reasoning
        ✓ Large Language Models Reasoning

Recent News

Oct 16, 2025 Our paper “Learning How to Use Tools, Not Just When: Pattern-Aware Tool-Integrated Reasoning” got accepted in *MATH-AI Workshop, Neurips 2025.
Aug 20, 2025 One paper accepted in EMNLP 2025.

Featured Publications

Check out Google Scholar for a full list of my publications.

  1. DRP: Distilled Reasoning Pruning with Skill-aware Step Decomposition for Efficient Large Reasoning Models
    Yuxuan Jiang, Dawei Li, and Francis Ferraro
    arXiv preprint, May 2025
  2. NeurIPS
    Learning How to Use Tools, Not Just When: Pattern-Aware Tool-Integrated Reasoning
    Ningning Xu, Yuxuan Jiang, and Shubhashis Roy Dipta
    In NeurIPS 2025 Workshop on Math-AI, Dec 2025
    MATH-AI Workshop