AI Paper Atlas

Research map / Vision-language and multimodal / Visual question answering

PromptCap: Prompt-Guided Image Captioning for VQA with GPT-3

Yushi Hu, Hang Hua, Zhengyuan Yang, Weijia Shi, Noah A. Smith, Jiebo Luo

ICCV 2023 · Vision-language and multimodal › Visual question answering · approach: Universal VQA models · 203 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use question-aware PromptCap captions instead of generic OFA captions (ablation)GPT-3 in-context VQA · OK-VQA valaccuracy (%): 56.6 -> 60.4Improved
use PromptCap captions instead of generic OFA captions (ablation)GPT-3 in-context VQA · A-OKVQA val (direct answer)accuracy (%): 51.0 -> 56.3Improved
use PromptCap captions instead of generic OFA captions (ablation)GPT-3 in-context VQA · VQAv2 valaccuracy (%): 64.9 -> 74.1Improved
use GPT-3 (175B) instead of Flan-T5-XXL (11B) as the language model (ablation)PromptCap pipeline · OK-VQA valaccuracy (%): 42.0 -> 60.4Improved
use GPT-3 instead of Flan-T5-XXL (ablation)PromptCap pipeline · VQAv2 valaccuracy (%): 70.9 -> 74.1Improved
increase in-context examples from 1 to 32 (ablation)PromptCap + GPT-3 with random examples · OK-VQA valaccuracy (%): 46.5 -> 55.2Improved
retrieve in-context examples with CLIP instead of random (ablation)PromptCap + GPT-3 · OK-VQA val, n=32accuracy (%): 55.2 -> 60.4Improved
use PromptCap captions instead of generic captions on out-of-domain WebQA (ablation)GPT-3 8-shot · WebQA valFL*Acc: 33.5 -> 34.5Improved
use PromptCap instead of OFA-Cap for COCO ground-truth caption similarity (ablation)caption quality · COCO-GT captions CIDErCIDEr: 147.9 -> 150.1Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy