The FUSE command right now allows configuring a single normalizer that will be applied on each result set.

FROM books METADATA _id, _index, _score
| FORK (WHERE title:"Shakespeare" | SORT _score DESC)
       (WHERE semantic_title:"Shakespeare" | SORT _score DESC)
| FUSE LINEAR WITH { "weights": { "fork1": 0.7, "fork2": 0.3 }, "normalizer": "minmax" }

We would like to be able to configure a normalizer for each branch of results:

FROM books METADATA _id, _index, _score
| FORK (WHERE title:"Shakespeare" | SORT _score DESC)
       (WHERE semantic_title:"Shakespeare" | SORT _score DESC)
| FUSE LINEAR WITH { "weights": { "fork1": 0.7, "fork2": 0.3 }, "normalizers": { "fork1": "minmax", "fork2": "none" } }

this might be helpful for example when combining lexical + knn results.
the vector search scores could already reside between 0 and 1 and normalization could be skipped for vector results, but still needed for lexical ones.

this feature also brings us closer to parity with the linear retriever, where different normalizers can be applied for each sub retriever.