EPIC uses similarity-gated cross-modal prompt interactions in frozen CLIP to improve text-image classification accuracy while training about 1% of parameters.
Memory fusion network for multi-view sequential learning,
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
-
EPIC: Efficient Prompt Interaction for Text-Image Classification
EPIC uses similarity-gated cross-modal prompt interactions in frozen CLIP to improve text-image classification accuracy while training about 1% of parameters.