Bio
Cheng Xin is an Assistant Professor of Computer Science at California State University, Fresno. He received his Ph.D. in Computer Science from Purdue University under the supervision of Dr. Tamal K. Dey and was a postdoctoral researcher at Rutgers University advised by Dr. Jie Gao. His research develops trustworthy and theoretically grounded AI systems by connecting topology, geometry, algorithms, and multi-agent interaction. His work spans trustworthy and multi-agent AI, topological and geometric learning, non-Euclidean representations, 3D and video generative AI, and AI for Science, with publications in NeurIPS, ICML, CVPR, and SoCG.
Research Interests
Topological and Geometric Machine Learning
Topological Data Analysis
3D & Video Generative AI and Evaluation
Interpretable AI
Multi-Agent Safety & Society
AI for Science
Academic Appointments & Education
Assistant Professor of Computer Science
- Research on trustworthy and multi-agent AI, topological and geometric learning, and theoretical foundations for reliable learning systems
- Teaching mathematical foundations of computer science, formal languages, and automata theory
Postdoctoral Researcher, Advisor: Prof. Jie Gao
- Developed topological frameworks for interpretable AI: TopInG for graph learning on molecular benchmarks
- Developed methods for non-Euclidean representation learning: Neuc-MDS and Johnson-Lindenstrauss extensions with provable theoretical guarantees
- Contributed to large-scale benchmarks for 3D vision and novel view synthesis: DL3DV-10K dataset
- Designed algorithms for multi-agent learning in social settings
Ph.D. in Computer Science, Advisor: Prof. Tamal K. Dey
- Dissertation: Decomposition and Stability of Multiparameter Persistence Modules
Doctoral studies in Computer Science, Advisor: Prof. Tamal K. Dey
- Developed generalized persistence algorithms for multiparameter persistence modules
M.S. in Computer Science, Advisor: Prof. Xiaolei Huang
- Thesis: Machine Learning Techniques for Cervigram Image Analysis
- Research Focus: Medical image analysis, machine learning applications
B.Eng. in Software Engineering
Invited Talks
"TopInG: Topologically Interpretable Graph Learning via Persistent Rationale Filtration"
TDA Conference, University of Missouri · Nov 2025
"Understanding through Shape of Data: Topological Data Analysis for Interpretable AI"
Management Science & Information Systems Colloquium, Rutgers University · Oct 2024
"Exploring Representations Beyond Euclidean Geometry"
John Hopcroft Center Seminar, Shanghai Jiao Tong University · Jun 2024
"Multiparameter Persistence and Its Applications"
CS Theory Seminar, Rutgers University · Nov 2023
"Generalized persistence algorithm for decomposing multi-parameter persistence modules"
Applied Algebraic Topology Network Seminar · Jul 2020
Peer-Reviewed Publications
[P7] CVPR 2024 Lu Ling, Yichen Sheng, Zhi Tu, Wentian Zhao, Cheng Xin, Kun Wan, Lantao Yu, Qianyu Guo, Zixun Yu, Yawen Lu, Xuanmao Li, Xingpeng Sun, Rohan Ashok, Aniruddha Mukherjee, Hao Kang, Xiangrui Kong, Gang Hua, Tianyi Zhang, Bedrich Benes, Aniket Bera. DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Vision.
CVPR 2024, pp. 22160-22169.
† Authors listed alphabetically. · * Co-first author.
Submitted Manuscripts
ArXiv links identify preprint versions of the same manuscripts.
[S1] ICLR 2027 submission Cheng Xin, Chenqi Guo. PCBench: Physical Commonsense Benchmark for Agentic Video Evaluation and Generation. Submitted to ICLR 2027, Sep 2026.
[S2] ICLR 2027 submission Cheng Xin. The Cost of Eigenspace Restriction in Shared Attention Control. Submitted to ICLR 2027, Sep 2026.
[S3] ICLR 2027 submission Cheng Xin. Exact Minimum-Event Explanations under Localized Observation Uncertainty. Submitted to ICLR 2027, Sep 2026.
[S4] ITCS 2027 submission Chengyuan Deng, Jie Gao, Kuijun Liu, Kevin Lu, Feng Luo, Puheng Weng, Cheng Xin. Navigable Graphs for Low-Dimensional Hyperbolic Spaces. Submitted to ITCS 2027, Sep 2026.
[S5] ITCS 2027 submission Cheng Xin. Quantum Query Algorithms for the Constructive Diagonal Ramsey Theorem.
Submitted to ITCS 2027, Sep 2026. arXiv:2609.05812 (preprint).
[S6] ITCS 2027 submission Cheng Xin. Quantum Query Complexity of Persistence Statistics in Graph Zigzags.
Submitted to ITCS 2027, Sep 2026. arXiv:2609.05813 (preprint).
[S7] ITCS 2027 submission Cheng Xin. Exact Kernel Transfer to Clique Complexes and the Hardness of Normalized Persistence.
Submitted to ITCS 2027, Sep 2026. arXiv:2610.00075 (preprint).
Other Preprint
[R1] Preprint 2021 Tamal K. Dey, Cheng Xin†. Rectangular Approximation and Stability of 2-parameter Persistence Modules.
Preprint, 2021. arXiv:2108.07429.
Teaching Experience
Instructor, CSCI 60: Foundations of Computer Science, Fresno State · Fall 2026.
Developed notes, slides, and assessments on sets, logic, and inference.
Instructor, CSCI 119: Formal Languages & Automata, Fresno State · Fall 2026.
Developed DFA lessons, proof-based problem sets, tagged PDFs, and LaTeX templates.
Lecturer, CS 513: Algorithms, Rutgers (graduate; 45 students) · Fall 2025.
TA, Data Structures & Algorithms (undergrad; 200 students) · Spring 2023.
TA, Computational Geometry (graduate; 30 students) · Fall 2020.
Professional Service
Area Chair, 2nd Annual Conference on Topology, Algebra, and Geometry in Data Science (TAG-DS 2026), August 2026.
Reviewer, ICLR 2027 (review service in 2026).
Reviewer, ICML, NeurIPS, SoCG.
Industrial Experience
Big Data Group · Redwood City, CA
- Large-scale machine learning on Spark
- Graph learning on relational database, attributes evaluation and selection, dataset compression
AWS Infrastructure Group · Seattle, WA
- Data management system for network messages supporting receiving, parsing, storing, and retrieving
References
Reference contact information; confidential letters are held in Interfolio Dossier.
Dr. Tamal K. Dey
Professor, Computer Science, Purdue University
tamaldey@purdue.edu
Dr. Jie Gao
Professor, Computer Science, Rutgers University
jg1555@cs.rutgers.edu
Dr. Feng Luo
Professor, Mathematics, Rutgers University
fluo@math.rutgers.edu
Updated October 18, 2026