A training-free PCA and LLM-based method shifts abstract fashion queries toward concrete caption embeddings, improving text-to-image retrieval in vision-language models.
Concreteness ratings for 40 thousand generally known en- glish word lemmas
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
-
Seeing the Abstract: Translating the Abstract Language for Vision Language Models
A training-free PCA and LLM-based method shifts abstract fashion queries toward concrete caption embeddings, improving text-to-image retrieval in vision-language models.