CoViP is a unified framework for contextualized visual personalization in VLMs that treats personalized image captioning as the core task, applies RL-based post-training and caption-augmented generation, and shows gains on diagnostic evaluations that rule out textual shortcuts plus downstream tasks.
yes" only when every concept appears in the query image. Carefully examine the images and output the final result only as
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Contextualized Visual Personalization in Vision-Language Models
CoViP is a unified framework for contextualized visual personalization in VLMs that treats personalized image captioning as the core task, applies RL-based post-training and caption-augmented generation, and shows gains on diagnostic evaluations that rule out textual shortcuts plus downstream tasks.