The stellar mass - black hole mass (M*-M_BH) scaling relation is key to understanding the co-evolution of galaxies and their central supermassive black holes (SMBHs). In this talk, I will report our recent work on the M*-M_BH relation investigated with type 1 AGN down to M_BH=10^7 Msun at z=2-2.5. Exploiting the deep and large-area spectroscopic survey of the Hobby-Eberly Telescope Dark Energy Experiment (HETDEX), we identify 66 type 1 AGN with M_BH ranging from 10^7 to 10^10 Msun, and estimate M* of the host galaxies with CIGALE SED fitting based on optical to near-infrared photometric data. We further assess the validity of SED fitting in two cases by host-nuclear decomposition performed through surface brightness profile fitting on spatially-resolved host galaxies with JWST/NIRCam data. We obtain the M*-M_BH relation covering the unexplored low-mass ranges of M_BH~10^7-10^8 Msun, and conduct forward modelling to account for the selection biases and observational uncertainties. The intrinsic M*-M_BH relation at z~2 has a moderate positive offset of 0.52\pm0.14 dex from the local relation, suggestive of more efficient black hole growth at higher redshift even in the low-mass regime of M_BH~10^7-10^8 Msun. Our M*-M_BH relation is inconsistent with the M_BH suppression at the low-M* regime predicted by recent hydrodynamic simulations at a 97% confidence level, suggesting that feedback in the low-mass systems may be weaker than those produced in hydrodynamic simulations.