Perhaps we will have to look for flowers before we assign this specimen to any Trichosanthes species. The crucial difference is fimbriate corolla of T. bracteata (Lam) Voight (syn T. palmata Roxb.) in which fruits are without costae, T. wallichiana similarly without any costae. Not much images are available for T. khasiana Kundo, which was created based on type by Hooker and Thoms. as indicated here
https://psimg.jstor.org/fsi/img/pdf/i0/10.5555/al.ap.specimen.k000102019_normal.pdf, distributed as var of T. multiloba in FBI
Costae are visible in only fruit seen, so it might be T. costata as suggested by Saroj ji, but as I mentioned above flowers are essential for confirmation.
My images on Flowers of India