Nobin Sarwar

About

𝗡𝗟𝗣 𝗣𝗵𝗗 𝘀𝘁𝘂𝗱𝗲𝗻𝘁 @ 𝗨𝗠𝗕𝗖 𝗖𝗦 working on 𝗽𝗼𝘀𝘁-𝘁𝗿𝗮𝗶𝗻𝗶𝗻𝗴 𝗼𝗽𝘁𝗶𝗺𝗶𝘇𝗮𝘁𝗶𝗼𝗻 and 𝘀𝘆𝘀𝘁𝗲𝗺-𝗹𝗲𝘃𝗲𝗹 𝗺𝗲𝘁𝗵𝗼𝗱𝘀 𝗳𝗼𝗿 𝗟𝗟𝗠𝘀 and 𝗺𝘂𝗹𝘁𝗶𝗺𝗼𝗱𝗮𝗹 𝗳𝗼𝘂𝗻𝗱𝗮𝘁𝗶𝗼𝗻 𝗺𝗼𝗱𝗲𝗹𝘀 (LLM safety, agentic AI). My research includes Multimodal Unlearning Survey (𝗔𝗖𝗟 𝟮𝟬𝟮𝟲 𝗙𝗶𝗻𝗱𝗶𝗻𝗴𝘀), FedMentor (𝗚𝗲𝗻𝗔𝗜𝟰𝗛𝗲𝗮𝗹𝘁𝗵 @ 𝗡𝗲𝘂𝗿𝗜𝗣𝗦 𝟮𝟬𝟮𝟱), and FilterRAG (𝗧𝟮𝗙𝗠 @ 𝗜𝗖𝗖𝗩 𝟮𝟬𝟮𝟱).

𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝗦𝗰𝗶𝗲𝗻𝘁𝗶𝘀𝘁 𝗜𝗻𝘁𝗲𝗿𝗻 @ 𝗟𝗟𝗡𝗟 (Summer 2026) — AI safety, protecting confidential data when using frontier models, scientific reasoning, and adversarial evaluation; featured in Meet LLNL Interns. Previously, MS @ 𝗨𝗧𝗥𝗚𝗩 𝗖𝗦; worked on privacy-preserving Federated Learning for biometrics and Differential Privacy, and contributed research ideas that supported securing an 𝗡𝗦𝗙-funded grant on Federated Learning for swarm robotics.

𝗢𝗽𝗲𝗻 𝘁𝗼 𝗰𝗼𝗹𝗹𝗮𝗯𝗼𝗿𝗮𝘁𝗶𝗼𝗻𝘀 on 𝗮𝗴𝗲𝗻𝘁𝗶𝗰 𝗿𝗲𝗮𝘀𝗼𝗻𝗶𝗻𝗴, 𝗽𝗼𝘀𝘁-𝘁𝗿𝗮𝗶𝗻𝗶𝗻𝗴, and 𝗟𝗟𝗠 𝗲𝘃𝗮𝗹𝘂𝗮𝘁𝗶𝗼𝗻 — and to 𝗺𝗲𝗻𝘁𝗼𝗿𝗶𝗻𝗴 undergrad and grad students new to research. Email me anytime.

𝗔𝗰𝗮𝗱𝗲𝗺𝗶𝗰 𝗽𝗮𝗴𝗲: smsnobin77.github.io

Employment

• Lawrence Livermore National Laboratory Research Scientist Intern
  2026 - 2026
• University of Maryland, Baltimore County Graduate Research Assistant
  2024 - Present
• The University of Texas Rio Grande Valley Graduate Research Assistant
  2021 - 2023

Education

• The University of Texas Rio Grande Valley MS in Computer Science
  2021 - 2023

Projects & Funding

Projects & funding information is unavailable.

Publications (4)

• Multimodal Unlearning Across Vision, Language, Video, and Audio: Survey of Methods, Datasets, and Benchmarks
  2026 DOI: 10.36227/techrxiv.176945748.88280394/v1
  +Save
• Differentially Private Data Publication with Multi-level Data Utility
  High-Confidence Computing 2022 DOI: 10.1016/j.hcc.2022.100049
  +Save
• Differential Privacy in Privacy-Preserving Big Data and Learning: Challenge and Opportunity
  2022 DOI: 10.1007/978-3-030-96057-5_3
  +Save
• Facial Spoof Detection Using Support Vector Machine
  Advances in Intelligent Systems and Computing 2020 DOI: 10.1007/978-981-33-4673-4_50
  +Save