AI Paper Atlas

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

Knowledge-Based Embodied Question Answering

Sinan Tan, Mengmeng Ge, Di Guo, Huaping Liu, Fuchun Sun

TPAMI 2023 · Vision-language and multimodal › Visual question answering · approach: Attention and prior-robust VQA · 45 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 the knowledge graph to the program synthesizer (ablation)Ours · K-EQA testoverall accuracy: 31.8 -> 41.3Improved
add planner-selected relevant-region exploration instead of full scan (ablation)Ours · K-EQA testoverall path length: 131.1 -> 82.7Improved
add planner-selected region exploration instead of full scan (ablation)Ours · K-EQA testoverall accuracy: 42.0 -> 41.3No clear effect
replace predicted segmentation with ground truth segmentation (ablation)Ours · K-EQA testoverall accuracy: 41.3 -> 67.5Improved
replace voxel scene graph with ground truth scene graph (ablation)Ours · K-EQA testoverall accuracy: 41.3 -> 90.3Improved
replace predicted segmentation with ground truth under full scan (ablation)Ours + FS · K-EQA testoverall accuracy: 42.0 -> 70.0Improved
evaluate on K-EQA Extension with logic connectors instead of K-EQAOurs · K-EQA vs Extensionoverall accuracy: 41.3 -> 37.1Hurt
increase agents from 1 to 5Ours · K-EQA multi-agentspeedup: 1.00 -> 2.32Improved
increase agents from 1 to 5Ours · K-EQA multi-agentaccuracy: 41.3 -> 40.4No clear effect
increase turns from 1 to 10 with scene graph memoryOurs · K-EQA multi-turnpath length: 82.7 -> 12.5Improved
increase turns from 1 to 10 with scene graph memoryOurs · K-EQA multi-turnaccuracy: 41.3 -> 42.7No clear effect

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