Jul 2026
Calibrate Before Reason: Robust Visual Token Reduction against Semantic Drift in VLMs
CaRe is proposed, a training-free robust framework that calibrates compact visual representations before reasoning to preserve semantic fidelity in VLMs and outperforms state-of-the-art token reduction baselines.
Jiasheng Li, Zhong Ji, Yan Zhang et al.
· arXiv.org · 0 citations