works on
efficient inference 1model calibration 1semantic drift 1vision-language models 1visual token reduction 1

From the 1 of 1 linked paper with an AI index.

collaborators
Showing cs.CVShow all

1 paper · 1 filter

cs.CV2026

Calibrate Before Reason: Robust Visual Token Reduction against Semantic Drift in VLMs

Jiasheng Li, Zhong Ji, Yan Zhang +1

The paper proposes CaRe, a training‑free method that calibrates compact visual representations before reasoning to keep semantic consistency when reducing visual tokens in large vi…