Test-Time Low Rank Adaptation via Confidence Maximization for Zero-Shot Generalization of Vision-Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Imam, Raza, Gani, Hanan, Huzaifa, Muhammad, Nandakumar, Karthik
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916332683919360
author Imam, Raza
Gani, Hanan
Huzaifa, Muhammad
Nandakumar, Karthik
author_facet Imam, Raza
Gani, Hanan
Huzaifa, Muhammad
Nandakumar, Karthik
contents The conventional modus operandi for adapting pre-trained vision-language models (VLMs) during test-time involves tuning learnable prompts, ie, test-time prompt tuning. This paper introduces Test-Time Low-rank adaptation (TTL) as an alternative to prompt tuning for zero-shot generalization of large-scale VLMs. Taking inspiration from recent advancements in efficiently fine-tuning large language models, TTL offers a test-time parameter-efficient adaptation approach that updates the attention weights of the transformer encoder by maximizing prediction confidence. The self-supervised confidence maximization objective is specified using a weighted entropy loss that enforces consistency among predictions of augmented samples. TTL introduces only a small amount of trainable parameters for low-rank adapters in the model space while keeping the prompts and backbone frozen. Extensive experiments on a variety of natural distribution and cross-domain tasks show that TTL can outperform other techniques for test-time optimization of VLMs in strict zero-shot settings. Specifically, TTL outperforms test-time prompt tuning baselines with a significant improvement on average. Our code is available at at https://github.com/Razaimam45/TTL-Test-Time-Low-Rank-Adaptation.
format Preprint
id arxiv_https___arxiv_org_abs_2407_15913
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Test-Time Low Rank Adaptation via Confidence Maximization for Zero-Shot Generalization of Vision-Language Models
Imam, Raza
Gani, Hanan
Huzaifa, Muhammad
Nandakumar, Karthik
Computer Vision and Pattern Recognition
The conventional modus operandi for adapting pre-trained vision-language models (VLMs) during test-time involves tuning learnable prompts, ie, test-time prompt tuning. This paper introduces Test-Time Low-rank adaptation (TTL) as an alternative to prompt tuning for zero-shot generalization of large-scale VLMs. Taking inspiration from recent advancements in efficiently fine-tuning large language models, TTL offers a test-time parameter-efficient adaptation approach that updates the attention weights of the transformer encoder by maximizing prediction confidence. The self-supervised confidence maximization objective is specified using a weighted entropy loss that enforces consistency among predictions of augmented samples. TTL introduces only a small amount of trainable parameters for low-rank adapters in the model space while keeping the prompts and backbone frozen. Extensive experiments on a variety of natural distribution and cross-domain tasks show that TTL can outperform other techniques for test-time optimization of VLMs in strict zero-shot settings. Specifically, TTL outperforms test-time prompt tuning baselines with a significant improvement on average. Our code is available at at https://github.com/Razaimam45/TTL-Test-Time-Low-Rank-Adaptation.
title Test-Time Low Rank Adaptation via Confidence Maximization for Zero-Shot Generalization of Vision-Language Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.15913