Prompt Tuning for Generative Multimodal Pretrained Models

4 Aug 2022  ·  Hao Yang, Junyang Lin, An Yang, Peng Wang, Chang Zhou, Hongxia Yang ·

Prompt tuning has become a new paradigm for model tuning and it has demonstrated success in natural language pretraining and even vision pretraining. In this work, we explore the transfer of prompt tuning to multimodal pretraining, with a focus on generative multimodal pretrained models, instead of contrastive ones. Specifically, we implement prompt tuning on the unified sequence-to-sequence pretrained model adaptive to both understanding and generation tasks. Experimental results demonstrate that the light-weight prompt tuning can achieve comparable performance with finetuning and surpass other light-weight tuning methods. Besides, in comparison with finetuned models, the prompt-tuned models demonstrate improved robustness against adversarial attacks. We further figure out that experimental factors, including the prompt length, prompt depth, and reparameteratization, have great impacts on the model performance, and thus we empirically provide a recommendation for the setups of prompt tuning. Despite the observed advantages, we still find some limitations in prompt tuning, and we correspondingly point out the directions for future studies. Codes are available at \url{}

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Captioning COCO Captions Prompt Tuning BLEU-4 41.81 # 10
METEOR 31.51 # 6
CIDER 141.4 # 15
SPICE 24.42 # 13
Visual Entailment SNLI-VE test Prompt Tuning Accuracy 90.12 # 2
Visual Entailment SNLI-VE val Prompt Tuning Accuracy 90.04 # 2
Visual Question Answering (VQA) VQA v2 test-std Prompt Tuning overall 78.53 # 11


No methods listed for this paper. Add relevant methods here