FREE: Fast and Robust Vision Language Models with Early Exits

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bajpai, Divya Jyoti, Hanawal, Manjesh Kumar
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913884814704640
author Bajpai, Divya Jyoti
Hanawal, Manjesh Kumar
author_facet Bajpai, Divya Jyoti
Hanawal, Manjesh Kumar
contents In recent years, Vision-Language Models (VLMs) have shown remarkable performance improvements in Vision-Language tasks. However, their large size poses challenges for real-world applications where inference latency is a concern. To tackle this issue, we propose employing Early Exit (EE) strategies in VLMs. However, training exit classifiers in VLMs is challenging, particularly with limited labeled training data. To address this, we introduce FREE, an adversarial training approach within a GAN-based framework. Here, each exit consists of a transformer layer and a classifier. The transformer layer is adversarially trained to produce feature representations similar to the final layer, while a feature classifier serves as the discriminator. Our method focuses on performing input-adaptive inference that increases inference speed with minimal drop in performance. Experimental results demonstrate the effectiveness of our approach in enhancing accuracy and model robustness by mitigating overthinking and the phenomenon of mid-crisis that we highlight. We experimentally validate that our method speeds up the inference process by more than 1.51x while retaining comparable performance. The source code is available at https://github.com/Div290/FREE.
format Preprint
id arxiv_https___arxiv_org_abs_2506_06884
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FREE: Fast and Robust Vision Language Models with Early Exits
Bajpai, Divya Jyoti
Hanawal, Manjesh Kumar
Machine Learning
Computer Vision and Pattern Recognition
In recent years, Vision-Language Models (VLMs) have shown remarkable performance improvements in Vision-Language tasks. However, their large size poses challenges for real-world applications where inference latency is a concern. To tackle this issue, we propose employing Early Exit (EE) strategies in VLMs. However, training exit classifiers in VLMs is challenging, particularly with limited labeled training data. To address this, we introduce FREE, an adversarial training approach within a GAN-based framework. Here, each exit consists of a transformer layer and a classifier. The transformer layer is adversarially trained to produce feature representations similar to the final layer, while a feature classifier serves as the discriminator. Our method focuses on performing input-adaptive inference that increases inference speed with minimal drop in performance. Experimental results demonstrate the effectiveness of our approach in enhancing accuracy and model robustness by mitigating overthinking and the phenomenon of mid-crisis that we highlight. We experimentally validate that our method speeds up the inference process by more than 1.51x while retaining comparable performance. The source code is available at https://github.com/Div290/FREE.
title FREE: Fast and Robust Vision Language Models with Early Exits
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.06884