Minh Nhat VuCNLS Postdoctoral Research Associate
T-1
Office: TA-3, Bldg 1690, Room 125
Mail Stop: B258
Email: mvu@lanl.gov
Research highlightI enjoy exploring and developing rigorous mathematical tools to address practical challenges in computer science, with a focus on trustworthy machine learning. My research spans three main areas: explainability, adversarial robustness, and data security.
Talks at CNLS:
|  | Educational Background/Employment:- B.S. (2015) Electrical and Electronics Engineering, Hanoi University of Science and Technology
- M.S. (2018) Electrical Engineering, University of Akron
- Ph.D. (2023) Computer Science, University of Florida
Research Interests:
- Machine Learning
- Information Theory
Google ScholarSelected Recent Publications:
- M. Vu; G. Zollicoffer; B. Nebgen; J. Castorena; B. Alexandrov; M. Bhattarai. LoRID: Low-Rank Iterative Diffusion for Adversarial Purification (2024) 10.48550/arXiv.2403.04784
- M. Bhattarai; R. Barron; M.E. Eren; M. Vu; V. Grantcharov; I. Boureima; V. Stanev; C. Matuszek; V. Valtchinov; K.Ø. Rasmussen; B. Alexandrov. HEAL: Hierarchical Embedding Alignment Loss for Improved Retrieval and Representation Learning (2025) 10.48550/arXiv.2412.04661
- M. Bhattarai; M. Vu; J. E. Santos; I. Boureima; D. O’ Malley. Enhancing Cross-Language Code Translation via Task-Specific Embedding Alignment in Retrieval-Augmented Generation (2025) 10.48550/arXiv.2412.05159
- R. Colman; M. Vu; M. Bhattarai, M. Ma; H. Viswanathan; D. O’Malley; J. E. Santos. PatchFinder: Leveraging Visual Language Models for Accurate Information Retrieval using Model Uncertainty (2024) 10.48550/arXiv.2412.02886
- M. Vu; T. Nguyen; My T. Thai. Analysis of Privacy Leakage in Federated Large Language Models (2024) 10.48550/arXiv.2403.04784
CNLS Publications:
- Vu, Minh Nhat; Tran, Brian Kha; Shah, Syed Alamdar Hussain; Zollicoffer, Geigh Marcel, et al. HALLUFIELD: DETECTING LLM HALLUCINATIONS VIA FIELD-THEORETIC MODELING. 2025 10.48550/arXiv.2509.10753
- Vu, Minh Nhat; Thai, My Tra; Nguyen, Quan Minh; Nguyen, Truc. Theoretically Unmasking Inference Attacks Against LDP-Protected Clients in Federated Vision Models. 2025 https://cnls.lanl.gov/Externals/pubs/LA-UR-25-24697.pdf
- Zollicoffer, Geigh Marcel; Vu, Minh Nhat; Bhattarai, Manish. Surprise to Hallucination: Detection of Failures in Vision-Language Models via Multi-token Aggregation. 2025 https://cnls.lanl.gov/Externals/pubs/LA-UR-25-24783.pdf
- Kunwar, Pradip; Vu, Minh Nhat; Gupta,Maanak; Abdelsalam,Mahmoud, et al. TT-LoRA MoE: Unifying Parameter-Efficient Fine-Tuning and Sparse Mixture-of-Experts. 2025 https://cnls.lanl.gov/Externals/pubs/LA-UR-25-23693.pdf
- Vu, Minh Nhat; Zollicoffer, Geigh Marcel; Mai,Huy; Nebgen, Benjamin Tyler, et al. Topological Signatures of Adversaries in Multimodal Alignments. 2025 https://cnls.lanl.gov/Externals/pubs/LA-UR-25-20561.pdf
- Bhattarai, Manish; Barron, Ryan Calvin; Eren, Maksim Ekin; Vu, Minh Nhat, et al. HEAL: Hierarchical Embedding Alignment Loss for Improved Retrieval and Representation Learning. 2024 https://cnls.lanl.gov/Externals/pubs/LA-UR-24-32672.pdf
- Bhattarai, Manish; Vu, Minh Nhat; Estrada Santos, Javier Andres; Djibrilla Boureima, Ismael, et al. Enhancing Cross-Language Code Translation via Task-Specific Embedding Alignment in Retrieval-Augmented Generation. 2024 https://cnls.lanl.gov/Externals/pubs/LA-UR-24-31097.pdf
- Zollicoffer, Geigh Marcel; Vu, Minh Nhat; Nebgen, Benjamin Tyler; Castorena Martinez, Juan Enrique, et al. LoRID: Low-Rank Iterative Diffusion for Adversarial Purification. 2024 10.1609/aaai.v39i21.34472
- Estrada Santos, Javier Andres; Colman, Roman James; Vu, Minh Nhat; Bhattarai, Manish, et al. PatchFinder: Improving OCR Accuracy in Visual Language Models using Model Uncertainty. 2024 10.48550/arXiv.2412.02886
- Vu, Minh Nhat; Bhattarai, Manish; Nebgen, Benjamin Tyler; Skau, Erik West, et al. LaFA: Latent Feature Attacks on Non-negative Matrix Factorization. 2024 https://cnls.lanl.gov/Externals/pubs/LA-UR-24-26951.pdf
|
