I am working on my thesis about option pricing models beyond classical Black-Scholes Model by looking for some recent innovations on mathematical processes behind the pricing structures. By that I mean doing some essential modification on traditional GBM framework or even replaced it with other mathematical processes, such as Levy's process (or some novel stuff). So far, I have found Neural SDEs to be a good starting point to this topic, since it replaces the fixed parameters in models like BSM and Heston's volatility model by neural networks. But I wonder if there are other scholar work introducing new mathematical frameworks of pricing model?
Any help is appreciated. Hopefully I could get some advice from researchers or pricing engineers as an undergraduate...
