About

Tengfei Ma is an Assistant Professor in the Department of Biomedical Informatics at Stony Brook University. His research focuses on machine learning, natural language processing, and artificial intelligence applications in healthcare. He is based in MART 7M-0810, Stony Brook, NY, and can be contacted at Tengfei.Ma@stonybrookmedicine.edu.

Research topics

• Computer Science
• Artificial Intelligence
• Machine Learning
• Data Mining
• Theoretical computer science
• Biology
• Bioinformatics
• Mathematics
• Pharmacology
• Geometry
• Computational biology
• Combinatorics
• Internal medicine
• Medicine

Selected publications

• An automatic parking decision framework based on interactive perception prediction and multiple strategy planning

  Robotics and Autonomous Systems · 2026-05-03

  article
  PublisherDOI

• Wasserstein Graph Neural Networks for Graphs With Missing Attributes

  IEEE Transactions on Pattern Analysis and Machine Intelligence · 2025-05-08 · 1 citations

  article

  Missing node attributes pose a common problem in real-world graphs, impacting the performance of graph neural networks' representation learning. Existing GNNs often struggle to effectively leverage incomplete attribute information, as they are not specifically designed for graphs with missing attributes. To address this issue, we propose a novel node representation learning framework called Wasserstein Graph Neural Network (WGNN). Our approach aims to maximize the utility of limited observed attribute information and account for uncertainty caused by missing values. We achieve this by representing nodes as low-dimensional distributions obtained through attribute matrix decomposition. Additionally, we enhance representation expressiveness by introducing a unique message-passing schema that aggregates distributional information from neighboring nodes in the Wasserstein space. We evaluate the performance of WGNN in node classification tasks using both synthetic and real-world datasets under two missing-attribute scenarios. Moreover, we demonstrate the applicability of WGNN in recovering missing values and tackling matrix completion problems, specifically in graphs involving users and items. Experimental results on both tasks convincingly demonstrate the superiority of our proposed method.

  PublisherDOI

• Dual-Pathway Fusion of EHRs and Knowledge Graphs for Predicting Unseen Drug-Drug Interactions

  ArXiv.org · 2025-11-10

  preprintOpen accessSenior author

  Drug-drug interactions (DDIs) remain a major source of preventable harm, and many clinically important mechanisms are still unknown. Existing models either rely on pharmacologic knowledge graphs (KGs), which fail on unseen drugs, or on electronic health records (EHRs), which are noisy, temporal, and site-dependent. We introduce, to our knowledge, the first system that conditions KG relation scoring on patient-level EHR context and distills that reasoning into an EHR-only model for zero-shot inference. A fusion "Teacher" learns mechanism-specific relations for drug pairs represented in both sources, while a distilled "Student" generalizes to new or rarely used drugs without KG access at inference. Both operate under a shared ontology (set) of pharmacologic mechanisms (drug relations) to produce interpretable, auditable alerts rather than opaque risk scores. Trained on a multi-institution EHR corpus paired with a curated DrugBank DDI graph, and evaluated using a clinically aligned, decision-focused protocol with leakage-safe negatives that avoid artificially easy pairs, the system maintains precision across multi-institutuion test data, produces mechanism-specific, clinically consistent predictions, reduces false alerts (higher precision) at comparable overall detection performance (F1), and misses fewer true interactions compared to prior methods. Case studies further show zero-shot identification of clinically recognized CYP-mediated and pharmacodynamic mechanisms for drugs absent from the KG, supporting real-world use in clinical decision support and pharmacovigilance.

  PublisherOA PDFDOI

• Bank Credit and Trade Credit: A New Perspective from Bank Regulatory Penalties

  SSRN Electronic Journal · 2025-01-01

  preprintOpen access
  PublisherDOI

• Uncovering LLM-Generated Code: A Zero-Shot Synthetic Code Detector via Code Rewriting

  Proceedings of the AAAI Conference on Artificial Intelligence · 2025-04-11 · 6 citations

  articleOpen access

  Large Language Models (LLMs) have demonstrated remarkable proficiency in generating code. However, the misuse of LLM-generated (synthetic) code has raised concerns in both educational and industrial contexts, underscoring the urgent need for synthetic code detectors. Existing methods for detecting synthetic content are primarily designed for general text and struggle with code due to the unique grammatical structure of programming languages and the presence of numerous ``low-entropy'' tokens. Building on this, our work proposes a novel zero-shot synthetic code detector based on the similarity between the original code and its LLM-rewritten variants. Our method is based on the observation that differences between LLM-rewritten and original code tend to be smaller when the original code is synthetic. We utilize self-supervised contrastive learning to train a code similarity model and evaluate our approach on two synthetic code detection benchmarks. Our results demonstrate a significant improvement over existing SOTA synthetic content detectors, delivering notable gains in both performance and robustness on the APPS and MBPP benchmarks.

  PublisherOA PDFDOI

• Bank Credit and Trade Credit: A New Perspective from Bank Regulatory Penalties

  SSRN Electronic Journal · 2025-01-01

  preprintOpen access
  PublisherDOI

• Multimodal collaborative framework for complex parking scenarios: multimodal perception fusion, curvature‑continuous trajectory planning and hierarchical robust tracking control

  Expert Systems with Applications · 2025-09-02

  article
  PublisherDOI

• Predicting Drug–Drug Interaction via Dual-Drug Visual Representation

  Journal of Chemical Information and Modeling · 2025-09-26 · 2 citations

  articleCorresponding

  Drug-drug interaction (DDI) prediction is essential for ensuring medication safety and therapeutic efficacy. While existing models often rely on chemical descriptors or molecular graphs, they tend to overlook the rich spatial and structural cues embedded in visual molecules. To address this issue, we propose DDVR-DDI, a novel vision-based framework that predicts DDIs by encoding drug pairs as a single fused molecular image, enabling direct modeling of their potential interaction interface. To enhance representation learning of visual drug pairs, we introduce a two-stage self-supervised pretraining strategy: a position-invariant contrastive task improves understanding of certain drug pairs in different spatial variations, while a jigsaw puzzle task encourages fine-grained structural understanding. Additionally, we develop a multiexpert voting mechanism, where multiple models analyze distinct augmented views of each drug pair to boost prediction accuracy and stability through ensemble inference. Extensive experiments on benchmark DDI data sets show that our model achieves state-of-the-art performance. To further interpret its predictions, we employ Grad-CAM visualizations and conduct multiple experiments to validate the stability and interpretability of the model; furthermore, we conduct a case study on Ritonavir inhibition of CYP3A, revealing that our model consistently highlights chemically significant substructures. These findings underscore the potential of image-based modeling for both accurate prediction and mechanistic insight in drug interaction research.

  PublisherDOI

• Study on flow control strategy of a fuel valve directly-driven by two-phase hybrid stepper motor

  Flow Measurement and Instrumentation · 2025-07-16 · 1 citations

  articleSenior author
  PublisherDOI

• Enhancing Graph Representation Learning with Localized Topological Features

  ArXiv.org · 2025-01-15

  preprintOpen access

  Representation learning on graphs is a fundamental problem that can be crucial in various tasks. Graph neural networks, the dominant approach for graph representation learning, are limited in their representation power. Therefore, it can be beneficial to explicitly extract and incorporate high-order topological and geometric information into these models. In this paper, we propose a principled approach to extract the rich connectivity information of graphs based on the theory of persistent homology. Our method utilizes the topological features to enhance the representation learning of graph neural networks and achieve state-of-the-art performance on various node classification and link prediction benchmarks. We also explore the option of end-to-end learning of the topological features, i.e., treating topological computation as a differentiable operator during learning. Our theoretical analysis and empirical study provide insights and potential guidelines for employing topological features in graph learning tasks.

  PublisherOA PDFDOI

Frequent coauthors

• Cao Xiao

  27 shared

• Lingfei Wu

  26 shared

• Shouling Ji

  Zhejiang University

  22 shared

• Jimeng Sun

  17 shared

• Swarnadeep Saha

  16 shared

• V. Lakshma Reddy

  University of Sydney

  16 shared

• Rishi Arora

  16 shared

• Chul Sung

  16 shared

Labs

• Computer Vision LabPI

Education

• Ph.D., Computer Science

  University of California, Los Angeles

  2008

• M.S., Computer Science

  University of California, Los Angeles

  2003

• B.S., Computer Science

  University of Science and Technology of China

  2001

Awards & honors

• ISWC 2021 Best Paper Award of Research Track
• IBM Outstanding Research Accomplishment 2019
• IBM Outstanding Research Accomplishment 2022