Injection Without Distortion: Geometrically Constrained Knowledge Enhancement for Vision-Language Models

Zhongze Wu; Xiu Su; Feng Yang; Shan You; Jun Long; Yueyi Luo

AAAI 2026technical0 citations

Injection Without Distortion: Geometrically Constrained Knowledge Enhancement for Vision-Language Models

Zhongze Wu, Xiu Su, Feng Yang, Shan You, Jun Long, Yueyi Luo

Abstract

Vision-Language Models (VLMs) are widely used in tasks like Open-Vocabulary Object Detection and zero-shot Classification, owing to their powerful generalization. However, recent research reveals that VLMs exhibit significant performance instability when tasked with recognizing concepts at varying granularities (e.g., ``animal

BibTeX

@inproceedings{aaai2026_injectionwithout,
  title = {Injection Without Distortion: Geometrically Constrained Knowledge Enhancement for Vision-Language Models},
  author = {Zhongze Wu and Xiu Su and Feng Yang and Shan You and Jun Long and Yueyi Luo},
  booktitle = {AAAI 2026},
  year = {2026}
}