PromptCap: Prompt-Guided Image Captioning for VQA with GPT-3 | IEEE Conference Publication | IEEE Xplore