FVG-PT: Adaptive Foreground View-Guided Prompt Tuning for Vision-Language Models

TL;DR

This paper introduces FVG-PT, an adaptive prompt tuning method for vision-language models that guides visual attention towards foreground objects, improving task adaptation and addressing attention shift issues.

Contribution

FVG-PT proposes a novel foreground attention guidance module with a reliability gate, distillation, and calibration to enhance prompt tuning in VLMs.

Findings

FVG-PT improves foreground attention alignment across models.

Enhanced tuning results on multiple datasets.

Demonstrates compatibility with various backbone models.

Abstract

CLIP-based prompt tuning enables pretrained Vision-Language Models (VLMs) to efficiently adapt to downstream tasks. Although existing studies have made significant progress, they pay limited attention to changes in the internal attention representations of VLMs during the tuning process. In this paper, we attribute the failure modes of prompt tuning predictions to shifts in foreground attention of the visual encoder, and propose Foreground View-Guided Prompt Tuning (FVG-PT), an adaptive plug-and-play foreground attention guidance module, to alleviate the shifts. Concretely, FVG-PT introduces a learnable Foreground Reliability Gate to automatically enhance the foreground view quality, applies a Foreground Distillation Compensation module to guide visual attention toward the foreground, and further introduces a Prior Calibration module to mitigate generalization degradation caused by excessive focus on the foreground. Experiments on multiple backbone models and datasets show the effectiveness and compatibility of FVG-PT. Codes are available at: https://github.com/JREion/FVG-PT