Meta-Prompting for Automating Zero-shot Visual Recognition with LLMs

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Mirza, M. Jehanzeb, Karlinsky, Leonid, Lin, Wei, Doveh, Sivan, Micorek, Jakub, Kozinski, Mateusz, Kuehne, Hilde, Possegger, Horst
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909280777535488
author Mirza, M. Jehanzeb
Karlinsky, Leonid
Lin, Wei
Doveh, Sivan
Micorek, Jakub
Kozinski, Mateusz
Kuehne, Hilde
Possegger, Horst
author_facet Mirza, M. Jehanzeb
Karlinsky, Leonid
Lin, Wei
Doveh, Sivan
Micorek, Jakub
Kozinski, Mateusz
Kuehne, Hilde
Possegger, Horst
contents Prompt ensembling of Large Language Model (LLM) generated category-specific prompts has emerged as an effective method to enhance zero-shot recognition ability of Vision-Language Models (VLMs). To obtain these category-specific prompts, the present methods rely on hand-crafting the prompts to the LLMs for generating VLM prompts for the downstream tasks. However, this requires manually composing these task-specific prompts and still, they might not cover the diverse set of visual concepts and task-specific styles associated with the categories of interest. To effectively take humans out of the loop and completely automate the prompt generation process for zero-shot recognition, we propose Meta-Prompting for Visual Recognition (MPVR). Taking as input only minimal information about the target task, in the form of its short natural language description, and a list of associated class labels, MPVR automatically produces a diverse set of category-specific prompts resulting in a strong zero-shot classifier. MPVR generalizes effectively across various popular zero-shot image recognition benchmarks belonging to widely different domains when tested with multiple LLMs and VLMs. For example, MPVR obtains a zero-shot recognition improvement over CLIP by up to 19.8% and 18.2% (5.0% and 4.5% on average over 20 datasets) leveraging GPT and Mixtral LLMs, respectively
format Preprint
id arxiv_https___arxiv_org_abs_2403_11755
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Meta-Prompting for Automating Zero-shot Visual Recognition with LLMs
Mirza, M. Jehanzeb
Karlinsky, Leonid
Lin, Wei
Doveh, Sivan
Micorek, Jakub
Kozinski, Mateusz
Kuehne, Hilde
Possegger, Horst
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Prompt ensembling of Large Language Model (LLM) generated category-specific prompts has emerged as an effective method to enhance zero-shot recognition ability of Vision-Language Models (VLMs). To obtain these category-specific prompts, the present methods rely on hand-crafting the prompts to the LLMs for generating VLM prompts for the downstream tasks. However, this requires manually composing these task-specific prompts and still, they might not cover the diverse set of visual concepts and task-specific styles associated with the categories of interest. To effectively take humans out of the loop and completely automate the prompt generation process for zero-shot recognition, we propose Meta-Prompting for Visual Recognition (MPVR). Taking as input only minimal information about the target task, in the form of its short natural language description, and a list of associated class labels, MPVR automatically produces a diverse set of category-specific prompts resulting in a strong zero-shot classifier. MPVR generalizes effectively across various popular zero-shot image recognition benchmarks belonging to widely different domains when tested with multiple LLMs and VLMs. For example, MPVR obtains a zero-shot recognition improvement over CLIP by up to 19.8% and 18.2% (5.0% and 4.5% on average over 20 datasets) leveraging GPT and Mixtral LLMs, respectively
title Meta-Prompting for Automating Zero-shot Visual Recognition with LLMs
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2403.11755