SolBehave: Evaluating the Behavioural Correctness of LLM-Generated Smart Contracts
SolBehave is introduced, a benchmark for evaluating the correctness of model-generated contracts at scale, moving beyond ad hoc unit tests, and differential fuzzing is applied, a necessary condition for correctness and far from a sufficient one.
G. Delvecchio, Stefano Fantazzini, Giacomo Frisoni et al.
· 0 citations