crushing state-of-the-art in state inconsistency bug analysis

wrote a paper this semester that does the work of state-of-the-art dynamic SI detectors but without expensive multiplex symbolic execution and SMT. i'll break it down in simple terms without spoiling our (best) detector -- we need to sell a product after all.

smart-contract security detectors have historically taken 2 paths:

1. shallow static scans that catch surface-level issues only
2. heavy dynamic engines powered by MSE, fuzzing, or SMT solvers that formally prove valid issues but take forever

this semester, we decided to achieve perf present in (2), but with (1). and it worked.

here is a benchmark of our results compared to other tools on the market.

<< to be continued >>