Maheshwari, OmOmMaheshwariSingh, AishwaryaAishwaryaSinghMohapatra, Nihar RanjanNihar RanjanMohapatra2025-08-312025-08-312024-01-01[9798350371529]10.1109/EDTM58488.2024.105113112-s2.0-85193299515http://repository.iitgn.ac.in/handle/IITG2025/29120This work presents a computationally efficient approach to extract the compact model parameters without extensive training requirements. Bayesian optimization is employed in multiple stages to optimize different model parameters. The methodology is demonstrated on Virtual Source model (MVS 2.0), extended for Nanosheet FET and MoS<inf>2</inf> based 2DFET. Optimization function based on I-V characteristics slope, on and off currents ensures optimum fitting of global as well as local model parameters for diverse devices.false2DFET | Bayesian optimization | compact model | Nanosheet FET | Parameter extraction | Virtual sourceConference Paper20240cpConference Proceeding0