Visual Prompt Tuning
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 0
- 평가표
- 4
- 보존된 평가 기록
- 40
Visual Prompt Tuning(VPT) only introduces a small amount of task-specific learnable parameters into the input space while freezing the entire pre-trained Transformer backbone during downstream training. In practice, these additional parameters are simply prepended into the input sequence of each Transformer layer and learned together with a linear head during fine-tuning. VPT is especially effective in the low-data regime, and maintains its advantage across data scales. Finally, VPT is competitive for a range of Transformer scales and designs (ViTBase/Large/Huge, Swin). Put together, the results suggest that VPT is one of the most effective ways of adapting ever-growing vision backbones.
논문
이 태스크와 하위 분야의 관련 논문입니다. 실제 성능 비교는 평가 결과에서 확인하세요.
Towards Few-shot Out-of-Distribution Detection
TSP-Transformer: Task-Specific Prompts Boosted Transformer for Holistic Scene Understanding
Unlocking the Potential of Prompt-Tuning in Bridging Generalized and Personalized Federated Learning
Task-driven Prompt Evolution for Foundation Models
From Question to Exploration: Test-Time Adaptation in Semantic Segmentation?
VPA: Fully Test-Time Visual Prompt Adaptation
MVP: Meta Visual Prompt Tuning for Few-Shot Remote Sensing Image Scene Classification
Dynamic Visual Prompt Tuning for Parameter Efficient Transfer Learning
LoGoPrompt: Synthetic Text Images Can Be Good Visual Prompts for Vision-Language Models
Online Class Incremental Learning on Stochastic Blurry Task Boundary via Mask and Visual Prompt Tuning
SLPT: Selective Labeling Meets Prompt Tuning on Label-Limited Lesion Segmentation
E^2VPT: An Effective and Efficient Approach for Visual Prompt Tuning
DVPT: Dynamic Visual Prompt Tuning of Large Pre-trained Models for Medical Image Analysis
Uncertainty-aware State Space Transformer for Egocentric 3D Hand Trajectory Forecasting
Improving Visual Prompt Tuning for Self-supervised Vision Transformers
TransTIC: Transferring Transformer-based Image Compression from Human Perception to Machine Perception
Learning Disentangled Prompts for Compositional Image Synthesis
Do We Really Need a Large Number of Visual Prompts?
PVP: Pre-trained Visual Parameter-Efficient Tuning
Instance-aware Dynamic Prompt Tuning for Pre-trained Point Cloud Models
Gradient-Regulated Meta-Prompt Learning for Generalizable Vision-Language Models
Understanding Zero-Shot Adversarial Robustness for Large-Scale Models
Multitask Vision-Language Prompt Tuning
Unified Vision and Language Prompt Learning
Visual Prompt Tuning for Test-time Domain Adaptation
Visual Prompt Tuning for Generative Transfer Learning
Prompt-Matched Semantic Segmentation
Dual Modality Prompt Tuning for Vision-Language Pre-Trained Model
Visual Prompt Tuning
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.