AI Paper Atlas

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

Learning by Asking Questions for Knowledge-Based Novel Object Recognition

K. Uehara, Tatsuya Harada

IJCV 2024 · Vision-language and multimodal › Visual question answering · approach: Attention and prior-robust VQA · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

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
add knowledge acquired by question generation (confidence policy)CLIP-Ret. OC · Professional K-VQG merged (test set)Overall acc. (zero-shot): 38.9 -> 56.1Improved
add knowledge acquired by question generation (confidence policy)CLIP-Ret. OC · Professional K-VQG merged (test set)Novel acc. (zero-shot): 0.0 -> 38.3Improved
fine-tune OC on acquired knowledgeOurs · Professional K-VQG merged (test set)Overall acc.: 56.1 -> 62.0Improved
fine-tune OC on acquired knowledgeAll Conf. · Professional K-VQG merged (test set)Overall acc.: 56.2 -> 59.7Improved
fine-tune OC on acquired knowledgeAll Exp. · Professional K-VQG merged (test set)Known acc.: 57.1 -> 54.7Hurt
use mixed confirmation/exploration policy instead of all confirmationOurs vs All Conf. · Professional K-VQG merged (test set)Novel acc. (fine-tune): 45.3 -> 51.9Improved
use mixed confirmation/exploration policy instead of all confirmationOurs vs All Conf. · Professional K-VQG merged (test set)Overall acc. (zero-shot): 56.2 -> 56.1No clear effect
use confidence-based policy instead of random policyOurs vs Random Policy · Professional K-VQG merged (test set)Overall acc. (fine-tune): 53.2 ± 0.39 -> 62.0Improved
use confidence-based policy instead of all explorationOurs vs All Exp. · Professional K-VQG merged (test set)#valid questions: 1057 -> 9864Improved
use confidence-based policy instead of all explorationOurs vs All Exp. · Professional K-VQG merged (test set)Novel acc. (fine-tune): 28.4 -> 51.9Improved

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