Rust’s two most prominent type system features are lifetimes and traits. Traits aka type classes first appeared in Haskell and are not a feature of ML.

I would be interested to learn of features in Rust that came from ML and that aren’t features of Haskell - not counting strictness :-) I can think of a non-example: one of ML's distinguishing features is its functional (in the FP sense) module system; Haskell and Rust have more conventional module systems.

You could maybe argue that a Rust enum definition is more like an ML type definition than a Haskell data definition because the constructor parameters are tuples rather than curried, but it isn’t clear to me whether this is actual influence from ML, or Rust aiming to be more C-ish.

As I understand the history of algebraic data types and pattern matching, they first appeared in Hope, and from there they were added to ML, and on the separate branch of lazy functional languages they were inherited by Hope’s successors Miranda, Orwell, and Haskell.