imo a significant part of the problem is the insistence of the community on misnaming the paradigm as “multi-vector retrieval”, we called it late interaction for a reason
the problem isn’t in having one vector; the problem is in the scoring function!…
ok fine, just for fun and after years of realizing how much inertia exists even in 'fast-moving' ML of all places, i'm taking matters into my own (codex?) hands.. stay tuned
imo a significant part of the problem is the insistence of the community on misnaming the paradigm as “multi-vector retrieval”, we called it late interaction for a reason
the problem isn’t in having one vector; the problem is in the scoring function!…
ok fine, just for fun and after years of realizing how much inertia exists even in 'fast-moving' ML of all places, i'm taking matters into my own (codex?) hands.. stay tuned