College of Science and Mathematics

Cheng Xin

Assistant Professor

Computer Science | 559.278.4373 | Office: Science II 251

cxin@mail.fresnostate.edu

Education

Ph.D., Computer Science, Purdue University
M.S., Computer Science, Lehigh University
B.Eng., Software Engineering, Tongji University

Courses Taught

Fall 2026

CSCI 60: Foundations of Computer Science
CSCI 119: Introduction to Formal Languages and Automata

Personal website: jackal092927.github.io | Google Scholar

Cheng Xin is an Assistant Professor in the Department of Computer Science at California State University, Fresno. His research develops mathematical and algorithmic foundations for trustworthy artificial intelligence, connecting learning systems with topology, geometry, algorithms, and multi-agent interaction.

Before joining Fresno State, he was a postdoctoral researcher at Rutgers University, working with Professor Jie Gao. He earned his Ph.D. in Computer Science from Purdue University under the supervision of Professor Tamal K. Dey. His doctoral research focused on topological data analysis and the theory and algorithms of multiparameter persistent homology.

His work connects mathematical foundations with practical learning systems. A recurring goal is to build representations and algorithms that preserve the geometric, topological, and interaction structure of data while improving interpretability, robustness, and reliability. His research also spans data, benchmarks, and evaluation for 3D and video generative AI.

Trustworthy and multi-agent AI: Safety, robustness, and intervention in networked learning systems, including agent influence and reliable collective behavior.
Topological and geometric learning: Topological data analysis, multiparameter persistent homology, and geometric representations for interpretable learning. This includes non-Euclidean representations, dimensionality reduction, embeddings, and retrieval in hyperbolic and other geometric spaces.
3D and video generative AI: Generative modeling, datasets, benchmarks, and evaluation for 3D scenes and video, with an emphasis on visual quality, geometric consistency, and temporal coherence. This direction builds on collaborative work such as DL3DV-10K.
AI for science: Structure-aware learning for biological, molecular, and physical data, connecting geometric and topological representations with scientific modeling and analysis.

Recent Selected Publications

D-GRIL: End-To-End Topological Learning with 2-Parameter Persistence. SoCG 2026.
Locality Sensitive Hashing in Hyperbolic Space. SoCG 2026.
Johnson-Lindenstrauss Lemma Beyond Euclidean Geometry. NeurIPS 2025.
TopInG: Topologically Interpretable Graph Learning via Persistent Rationale Filtration. ICML 2025.
Neuc-MDS: Non-Euclidean Multidimensional Scaling Through Bilinear Forms. NeurIPS 2024.
Optimally Improving Cooperative Learning in a Social Setting. ICML 2024.
DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-Based 3D Vision. CVPR 2024.
GRIL: A 2-parameter Persistence Based Vectorization for Machine Learning. TAG-ML Workshop 2023, PMLR 221:313–333.
Generalized persistence algorithm for decomposing multiparameter persistence modules. Journal of Applied and Computational Topology 6(3):271–322, 2022.

Theses

Decomposition and Stability of Multiparameter Persistence Modules. Ph.D. dissertation, Purdue University, 2023.
Machine Learning Techniques for Cervigram Image Analysis. M.S. thesis, Lehigh University, 2016.

Collaborative Project

DL3DV-10K provides a large-scale real-world scene dataset and benchmark for 3D vision. Project website | Dataset and code

Selected Invited Talks

November 2025: “TopInG: Topologically Interpretable Graph Learning via Persistent Rationale Filtration.” Conference on Topological Data Analysis: Recent Developments and Applications, University of Missouri.
October 2024: “Understanding through Shape of Data: Topological Data Analysis for Interpretable AI.” Management Science and Information Systems Department Colloquium, Rutgers University.
June 2024: “Exploring Representations Beyond Euclidean Geometry.” John Hopcroft Center Seminar, Shanghai Jiao Tong University.
November 2023: “Multiparameter Persistence and Its Applications.” Theory Seminar, Department of Computer Science, Rutgers University.
July 2020: “Generalized persistence algorithm for decomposing multi-parameter persistence modules.” Applied Algebraic Topology Research Network Seminar.

Academic Service

Area Chair: Topology, Algebra, and Geometry in Data Science (TAG-DS), 2026.
Reviewer: ICML, ICLR, NeurIPS, and SoCG.

For a complete publication list, visit Google Scholar.