Preprint
Aug 2026
PatchGate: Narrowing the Verbalization Gap with Intrinsic Object Inventories in Frozen Vision-Language Models
PatchGate is proposed, a training-free framework that extracts prompt-free object evidence intrinsic to a frozen VLM before generation and uses it to narrow the gap between an intrinsic object set and final object mentions.
Jihyung Ko, Eunji Jung, Hyeongsub Kim et al.
· 0 citations