CROME: Cross-Modal Adapters for Efficient Multimodal LLM

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Ebrahimi, Sayna, Arik, Sercan O., Nama, Tejas, Pfister, Tomas
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911986422382592
author Ebrahimi, Sayna
Arik, Sercan O.
Nama, Tejas
Pfister, Tomas
author_facet Ebrahimi, Sayna
Arik, Sercan O.
Nama, Tejas
Pfister, Tomas
contents Multimodal Large Language Models (MLLMs) demonstrate remarkable image-language capabilities, but their widespread use faces challenges in cost-effective training and adaptation. Existing approaches often necessitate expensive language model retraining and limited adaptability. Additionally, the current focus on zero-shot performance improvements offers insufficient guidance for task-specific tuning. We propose CROME, an efficient vision-language instruction tuning framework. It features a novel gated cross-modal adapter that effectively combines visual and textual representations prior to input into a frozen LLM. This lightweight adapter, trained with minimal parameters, enables efficient cross-modal understanding. Notably, CROME demonstrates superior zero-shot performance on standard visual question answering and instruction-following benchmarks. Moreover, it yields fine-tuning with exceptional parameter efficiency, competing with task-specific specialist state-of-the-art methods. CROME demonstrates the potential of pre-LM alignment for building scalable, adaptable, and parameter-efficient multimodal models.
format Preprint
id arxiv_https___arxiv_org_abs_2408_06610
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CROME: Cross-Modal Adapters for Efficient Multimodal LLM
Ebrahimi, Sayna
Arik, Sercan O.
Nama, Tejas
Pfister, Tomas
Computer Vision and Pattern Recognition
Computation and Language
Machine Learning
Multimodal Large Language Models (MLLMs) demonstrate remarkable image-language capabilities, but their widespread use faces challenges in cost-effective training and adaptation. Existing approaches often necessitate expensive language model retraining and limited adaptability. Additionally, the current focus on zero-shot performance improvements offers insufficient guidance for task-specific tuning. We propose CROME, an efficient vision-language instruction tuning framework. It features a novel gated cross-modal adapter that effectively combines visual and textual representations prior to input into a frozen LLM. This lightweight adapter, trained with minimal parameters, enables efficient cross-modal understanding. Notably, CROME demonstrates superior zero-shot performance on standard visual question answering and instruction-following benchmarks. Moreover, it yields fine-tuning with exceptional parameter efficiency, competing with task-specific specialist state-of-the-art methods. CROME demonstrates the potential of pre-LM alignment for building scalable, adaptable, and parameter-efficient multimodal models.
title CROME: Cross-Modal Adapters for Efficient Multimodal LLM
topic Computer Vision and Pattern Recognition
Computation and Language
Machine Learning
url https://arxiv.org/abs/2408.06610