paper-with-me

홈 › Papers

When Does Visual Prompting Outperform Linear Probing for Vision-Language Models? A Likelihood Perspective

2024-09-03 · Hsi-Ai Tsao, Lei Hsiung, Pin-Yu Chen, Tsung-Yi Ho

Adapting pre-trained models to new tasks can exhibit varying effectiveness across datasets. Visual prompting, a state-of-the-art parameter-efficient transfer learning method, can significantly improve the performance of out-of-distribution tasks. On the other hand, linear probing, a standard transfer learning method, can sometimes become the best approach. We propose a log-likelihood ratio (LLR) approach to analyze the comparative benefits of visual prompting and linear probing. By employing the LLR score alongside resource-efficient visual prompts approximations, our cost-effective measure attains up to a 100-fold reduction in run time compared to full training, while achieving prediction accuracies up to 91%. The source code is available at https://github.com/IBM/VP-LLR.

📄 PDF Abstract BibTeX arXiv:2409.01821

Code (1)

ibm/vp-llr 공식 구현 pytorch

Tasks

Transfer LearningVisual Prompting

Similar Papers 제목 키워드 기반

Visual Access Boundaries in Vision-Language Model Reasoning

2026-07-14 · Hiroto Osaka, Shohei Taniguchi, Gouki Minegishi, Kai Yamashita 외 arxiv

Chain-of-Thought (CoT) prompting is widely used as a test-time scaling strategy for Vision-Language Models (VLMs), but it remains unclear what is extended when VLMs generate longer reasoning traces. We ask whether CoT re…

When More Is Less: A Systematic Analysis of Spatial and Commonsense Information for Visual Spatial Reasoning

2026-02-25 · Muku Akasaka, Soyeon Caren Han arxiv

Visual spatial reasoning (VSR) remains challenging for modern vision-language models (VLMs), despite advances in multimodal architectures. A common strategy is to inject additional information at inference time, such as …

Multimodal ReasoningSpatial Reasoning

Zero-Shot Visual Reasoning by Vision-Language Models: Benchmarking and Analysis

2024-08-27 · Aishik Nagar, Shantanu Jaiswal, Cheston Tan

Vision-language models (VLMs) have shown impressive zero- and few-shot performance on real-world visual question answering (VQA) benchmarks, alluding to their capabilities as visual reasoning engines. However, the benchm…

BenchmarkingLarge Language ModelQuestion AnsweringVisual Question Answering+3

Benchmarking Human and Automated Prompting in the Segment Anything Model

2024-10-29 · Jorge Quesada, Zoe Fowler, Mohammad Alotaibi, Mohit Prabhushankar 외

The remarkable capabilities of the Segment Anything Model (SAM) for tackling image segmentation tasks in an intuitive and interactive manner has sparked interest in the design of effective visual prompts. Such interest h…

BenchmarkingImage SegmentationSemantic SegmentationVisual Prompting

Unleashing the Power of Visual Prompting At the Pixel Level

2022-12-20 · Junyang Wu, Xianhang Li, Chen Wei, Huiyu Wang 외

This paper presents a simple and effective visual prompting method for adapting pre-trained models to downstream recognition tasks. Our method includes two key designs. First, rather than directly adding together the pro…

DiversityVisual Prompting