Zhongxing Zhang

Computer Science Ph.D. Candidate • University of Minnesota, Twin Cities
Department of Computer Science & Engineering • Advised by Prof. Jaideep Srivastava

I am a Ph.D. student in Computer Science at the University of Minnesota, Twin Cities, advised by Prof. Jaideep Srivastava. My research lies at the intersection of multimodal learning, large language models (LLMs), and computational social science, with a focus on combating misinformation in the wild. I build systems that jointly reason over text, images, and external knowledge to detect fake news and assess online information credibility.

Prior to UMN, I received my M.Eng. and B.Eng. from Shandong University of Finance and Economics, where I worked with Prof. Hui Liu on low-level computer vision, including image super-resolution and restoration.

Zhongxing Zhang

University of Minnesota

Minneapolis, MN 55455

zhan8889@umn.edu

News

2025Released BiMind — a dual-head reasoning model with Attention-Geometry Adapter for multimodal incorrect information detection.
2024Ongoing research on VLM-based multimodal fake news detection with retrieval-augmented external knowledge grounding.

Research Interests

  • Multimodal Misinformation Detection: Designing models that jointly reason over text and images to identify fake news, with applications to social media and cross-lingual settings.
  • Vision-Language Models (VLMs): Leveraging large pretrained models (CLIP, LLaVA, Qwen-VL) as reasoning engines for credibility assessment tasks.
  • Retrieval-Augmented Generation (RAG): Grounding model predictions with external knowledge to improve factual accuracy and interpretability.
  • Computational Social Science: Analyzing social behaviors, network dynamics, and information propagation to understand online misinformation ecosystems.
  • Low-Level Vision (Prior Work): Image super-resolution and restoration via low-rank priors and probabilistic nuclear norm minimization.

Education

  • Ph.D. in Computer Science, University of Minnesota, Twin Cities, 2027 (Expected)
  • M.Eng. in Digital Media Technology, Shandong University of Finance and Economics, 2023
  • B.Eng. in Computer Science and Technology, Shandong University of Finance and Economics, 2018

Selected Publications

  • Zhongxing Zhang, Emily K. Vraga, Jisu Huh, Jaideep Srivastava. BiMind: A Dual-Head Reasoning Model with Attention-Geometry Adapter for Incorrect Information Detection. (Accepted to ACL 2026) [GitHub]
  • Jiacheng Huang, Zhongxing Zhang, and Chao Su. LabelGenius: A Python Library for LLM-based Multimodal Content Labeling. (Accepted to CCR Special Issue on Generative AI) [DOI]
  • Zhongxing Zhang, Emily K. Vraga, Jisu Huh, Jaideep Srivastava. Learning to Reason Across Modalities with Modular Heads for Correctness-Oriented Information Detection. (Under preparation for ARR, May 2026)
  • Congrui Yin, Evan Wei, Zhongxing Zhang, Zaifu Zhan. PaperHelper: Knowledge-Based LLM QA Paper Reading Assistant. (arXiv preprint) [DOI]
  • Zhongxing Zhang, Hui Liu, and Qiang Guo. Single image super-resolution reconstruction via probability-inducing nuclear norm minimization. (In Preparation)
  • Zhongxing Zhang, Hui Liu, Qiang Guo. Image restoration using probability-inducing nuclear norm minimization. IEEE ICIP, Bordeaux, 2022. [DOI]
  • Zhongxing Zhang, Hui Liu, Qiang Guo, Yuxiu Lin. Super-resolution reconstruction using probability model combined with non-local low-rank prior. Journal of Computer-Aided Design & Computer Graphics, 2021. [DOI]
  • Zhongxing Zhang, Hui Liu, Qiang Guo, Yuxiu Lin. Single image super-resolution reconstruction using non-local low-rank prior. ML4CS, Guangzhou, 2020. [DOI]
  • Jingqi Song, Hui Liu, Yuxiu Lin, Zhongxing Zhang. Medical images super-resolution based on similarity learning. APCMHS, Seoul, 2019. [DOI]

Technical Skills

  • Languages: Python, MATLAB, C/C++, Java, SQL, JavaScript, HTML/CSS
  • ML / DL Frameworks: PyTorch, HuggingFace Transformers, PEFT (LoRA)
  • Multimodal & LLM Tools: CLIP, LLaVA, Qwen-VL, BLIP-2, vLLM, LangChain
  • Data & Visualization: Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn, Weights & Biases
  • Infrastructure: Git, Docker, SLURM (HPC), VS Code

Experience

  • Applied Scientist Intern, Adobe (May 2026 – August 2026)

  • Graduate Research Assistant, University of Minnesota (2023 – Present)
    • Designing multimodal fake news detection architectures combining VLMs, knowledge graphs, and cross-modal attention.
    • Developing retrieval-augmented pipelines for grounding LLM-based misinformation classifiers with external knowledge.
    • Investigating dual-head reasoning with geometry-aware adapters for improved credibility inference (BiMind).
  • Volunteer Teacher, China Western Volunteer Program
    • Taught Chemistry, English, Mathematics, and Computer Science in underserved communities.
    • Organized environmental protection campaigns and community outreach activities.

Honors & Awards

  • National Scholarship for Graduate Students
  • Postgraduate Research Scholarship
  • President’s Scholarship
  • Excellent Volunteer — China Western Volunteer Program