Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Uchiyama, Fumiya, Yanagi, Rintaro, Taniguchi, Shohei, Takashiro, Shota, Suzuki, Masahiro, Kataoka, Hirokatsu, Iwasawa, Yusuke, Matsuo, Yutaka
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2506.22881
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914621030400000
author Uchiyama, Fumiya
Yanagi, Rintaro
Taniguchi, Shohei
Takashiro, Shota
Suzuki, Masahiro
Kataoka, Hirokatsu
Iwasawa, Yusuke
Matsuo, Yutaka
author_facet Uchiyama, Fumiya
Yanagi, Rintaro
Taniguchi, Shohei
Takashiro, Shota
Suzuki, Masahiro
Kataoka, Hirokatsu
Iwasawa, Yusuke
Matsuo, Yutaka
contents Density ratio estimation is a core concept in statistical machine learning because it provides a unified mechanism for tasks such as importance weighting, divergence estimation, and likelihood-free inference, but its potential in vision and language models has not been fully explored. Modern vision-language encoders such as CLIP and SigLIP are trained with contrastive objectives that implicitly optimize log density ratios between joint and marginal image-text distributions, which implicitly learn similarity scores proportional to log density ratios. However, prior work has largely focused on their embedding utility, and the density-ratio structure induced by contrastive learning has not been systematically examined or exploited in multimodal applications. To address this gap, we reinterpret CLIP-style models as pretrained and general-purpose density ratio estimators and show that this perspective enables new algorithmic capabilities. We present a unified explanation of how contrastive objectives estimate density ratios and propose two practical applications: Importance Weight Learning and KL divergence estimation. Our Importance Weight Learning method requires only a single additional prompt and improves F1 scores by up to 7 points. We further show that CLIP-based density ratios support estimation of KL divergences that quantify how conditioning on an image or text alters the distribution of the other modality. Through qualitative examples and an N-gram analysis of captions, we find that these divergences capture semantic diversity and mode structure in multimodal data. Leveraging this property, we introduce a simple KL-guided data curation method that achieves performance competitive with LAION2B filtering.
format Preprint
id arxiv_https___arxiv_org_abs_2506_22881
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CLIP-like Model as a Foundational Density Ratio Estimator
Uchiyama, Fumiya
Yanagi, Rintaro
Taniguchi, Shohei
Takashiro, Shota
Suzuki, Masahiro
Kataoka, Hirokatsu
Iwasawa, Yusuke
Matsuo, Yutaka
Computer Vision and Pattern Recognition
Density ratio estimation is a core concept in statistical machine learning because it provides a unified mechanism for tasks such as importance weighting, divergence estimation, and likelihood-free inference, but its potential in vision and language models has not been fully explored. Modern vision-language encoders such as CLIP and SigLIP are trained with contrastive objectives that implicitly optimize log density ratios between joint and marginal image-text distributions, which implicitly learn similarity scores proportional to log density ratios. However, prior work has largely focused on their embedding utility, and the density-ratio structure induced by contrastive learning has not been systematically examined or exploited in multimodal applications. To address this gap, we reinterpret CLIP-style models as pretrained and general-purpose density ratio estimators and show that this perspective enables new algorithmic capabilities. We present a unified explanation of how contrastive objectives estimate density ratios and propose two practical applications: Importance Weight Learning and KL divergence estimation. Our Importance Weight Learning method requires only a single additional prompt and improves F1 scores by up to 7 points. We further show that CLIP-based density ratios support estimation of KL divergences that quantify how conditioning on an image or text alters the distribution of the other modality. Through qualitative examples and an N-gram analysis of captions, we find that these divergences capture semantic diversity and mode structure in multimodal data. Leveraging this property, we introduce a simple KL-guided data curation method that achieves performance competitive with LAION2B filtering.
title CLIP-like Model as a Foundational Density Ratio Estimator
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.22881