LSPT: Long-term Spatial Prompt Tuning for Visual Representation Learning

arXiv:2402.17406v1 Announce Type: cross
Abstract: Visual Prompt Tuning (VPT) techniques have gained prominence for their capacity to adapt pre-trained Vision Transformers (ViTs) to downstream visual tasks using specialized learnable tokens termed as prompts. Contemporary VPT methodologies, especially when employed with self-supervised vision transformers, often default to the introduction of new learnable prompts or gated prompt tokens predominantly sourced from the model's previous block. A pivotal oversight in such approaches is their failure to harness the potential of long-range previous blocks …

abstract adapt arxiv capacity cs.ai cs.cv cs.lg introduction long-term prompt prompts prompt tuning representation representation learning spatial tasks tokens transformers type vision vision transformers visual