Context-Adaptive Multi-Prompt Embedding with Large Language Models for Vision-Language Alignment

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kim, Dahun, Angelova, Anelia
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918115660529664
author Kim, Dahun
Angelova, Anelia
author_facet Kim, Dahun
Angelova, Anelia
contents We propose Context-Adaptive Multi-Prompt Embedding, a novel approach to enrich semantic representations in vision-language contrastive learning. Unlike standard CLIP-style models that rely on a single text embedding, our method introduces multiple structured prompts, each containing a distinct adaptive token that captures diverse semantic aspects of the input text. We leverage a pretrained LLM as the text encoder within the CLIP framework, processing all prompts jointly in a single forward pass. The resulting prompt embeddings are combined into a unified text representation, enabling semantically richer alignment with visual features. To further promote semantic diversity and representation quality, we incorporate a diversity regularization loss and a negation-aware loss, encouraging specialization across prompts and improving contrastive discrimination. Our method achieves consistent improvements on both image-text and video-text retrieval benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2508_02762
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Context-Adaptive Multi-Prompt Embedding with Large Language Models for Vision-Language Alignment
Kim, Dahun
Angelova, Anelia
Machine Learning
Artificial Intelligence
We propose Context-Adaptive Multi-Prompt Embedding, a novel approach to enrich semantic representations in vision-language contrastive learning. Unlike standard CLIP-style models that rely on a single text embedding, our method introduces multiple structured prompts, each containing a distinct adaptive token that captures diverse semantic aspects of the input text. We leverage a pretrained LLM as the text encoder within the CLIP framework, processing all prompts jointly in a single forward pass. The resulting prompt embeddings are combined into a unified text representation, enabling semantically richer alignment with visual features. To further promote semantic diversity and representation quality, we incorporate a diversity regularization loss and a negation-aware loss, encouraging specialization across prompts and improving contrastive discrimination. Our method achieves consistent improvements on both image-text and video-text retrieval benchmarks.
title Context-Adaptive Multi-Prompt Embedding with Large Language Models for Vision-Language Alignment
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2508.02762