Indexing Multimodal Language Models for Large-scale Image Retrieval

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Tharwat, Bahey, Kordopatis-Zilos, Giorgos, Suma, Pavel, Reid, Ian, Tolias, Giorgos
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915937878278144
author Tharwat, Bahey
Kordopatis-Zilos, Giorgos
Suma, Pavel
Reid, Ian
Tolias, Giorgos
author_facet Tharwat, Bahey
Kordopatis-Zilos, Giorgos
Suma, Pavel
Reid, Ian
Tolias, Giorgos
contents Multimodal Large Language Models (MLLMs) have demonstrated strong cross-modal reasoning capabilities, yet their potential for vision-only tasks remains underexplored. We investigate MLLMs as training-free similarity estimators for instance-level image-to-image retrieval. Our approach prompts the model with paired images and converts next-token probabilities into similarity scores, enabling zero-shot re-ranking within large-scale retrieval pipelines. This design avoids specialized architectures and fine-tuning, leveraging the rich visual discrimination learned during multimodal pre-training. We address scalability by combining MLLMs with memory-efficient indexing and top-$k$ candidate re-ranking. Experiments across diverse benchmarks show that MLLMs outperform task-specific re-rankers outside their native domains and exhibit superior robustness to clutter, occlusion, and small objects. Despite strong results, we identify failure modes under severe appearance changes, highlighting opportunities for future research. Our findings position MLLMs as a promising alternative for open-world large-scale image retrieval.
format Preprint
id arxiv_https___arxiv_org_abs_2604_13268
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Indexing Multimodal Language Models for Large-scale Image Retrieval
Tharwat, Bahey
Kordopatis-Zilos, Giorgos
Suma, Pavel
Reid, Ian
Tolias, Giorgos
Computer Vision and Pattern Recognition
Computation and Language
Information Retrieval
Multimodal Large Language Models (MLLMs) have demonstrated strong cross-modal reasoning capabilities, yet their potential for vision-only tasks remains underexplored. We investigate MLLMs as training-free similarity estimators for instance-level image-to-image retrieval. Our approach prompts the model with paired images and converts next-token probabilities into similarity scores, enabling zero-shot re-ranking within large-scale retrieval pipelines. This design avoids specialized architectures and fine-tuning, leveraging the rich visual discrimination learned during multimodal pre-training. We address scalability by combining MLLMs with memory-efficient indexing and top-$k$ candidate re-ranking. Experiments across diverse benchmarks show that MLLMs outperform task-specific re-rankers outside their native domains and exhibit superior robustness to clutter, occlusion, and small objects. Despite strong results, we identify failure modes under severe appearance changes, highlighting opportunities for future research. Our findings position MLLMs as a promising alternative for open-world large-scale image retrieval.
title Indexing Multimodal Language Models for Large-scale Image Retrieval
topic Computer Vision and Pattern Recognition
Computation and Language
Information Retrieval
url https://arxiv.org/abs/2604.13268