prompt tuning
Prompt tuning involves modifying the input prompts provided to large language models to elicit desired behaviors or outputs. It is a lighter and often more flexible method compared to full model fine-tuning, focusing on the interaction between the input and the model's pre-trained knowledge.
- MixPrompt: Efficient Mixed Prompting for Multimodal Semantic Segmentation
- Prompt Tuning Decision Transformers with Structured and Scalable Bandits
- Prompt Tuning Transformers for Data Memorization
- Test-Time Adaptive Object Detection with Foundation Model
- VIPAMIN: Visual Prompt Initialization via Embedding Selection and Subspace Expansion