The Blending Ratio Is Not Where the Performance Is: Diagnosing Prototype Blending for Few-Shot Adaptation of Vision-Language Models

This story is from 2026-08-26. It is preserved in the archive; the latest stories are on the live feed.

arXiv:2608.23634v1 Announce Type: new Abstract: Many few-shot adaptation methods for vision-language models classify with a convex combination of the zero-shot text prototype and the mean of the K labelled image features, with a single blending ratio routinely tuned on held-out labels, often on the…

Read the full story at arXiv cs.CV ↗

Timeline · 1 report

1. 2026-08-26 04:00 · arXiv cs.CV
   The Blending Ratio Is Not Where the Performance Is: Diagnosing Prototype Blending for Few-Shot Adaptation of Vision-Language Models