KeCO updates the visual feature keys of a small coreset with all leftover support images, and its diversity-based update outperforms retrieval from the five-times-larger full support set for LVLM in-context image classification.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Enhancing Multimodal In-Context Learning for Image Classification through Coreset Optimization
KeCO updates the visual feature keys of a small coreset with all leftover support images, and its diversity-based update outperforms retrieval from the five-times-larger full support set for LVLM in-context image classification.