Research map / Vision-language and multimodal / Visual question answering
IQA: Visual Question Answering in Interactive Environments
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use HIMN with YOLO detections instead of A3C with YOLO detections | IQA agent · IQA test — Existence | accuracy: 54.29 -> 63.39 | Helped |
| Use HIMN with YOLO detections instead of A3C with YOLO detections | IQA agent · IQA test — Counting | accuracy: 26.78 -> 35.89 | Helped |
| Use HIMN with YOLO detections instead of A3C with YOLO detections | IQA agent · IQA test — Spatial Relationship | accuracy: 55.36 -> 57.14 | Helped |
| Use HIMN with YOLO detections instead of A3C with ground-truth detections | IQA agent · IQA test — Spatial Relationship | accuracy: 66.2 -> 57.14 | Hurt |
| Add oracle object detections to HIMN instead of YOLO detections (ablation) | HIMN · IQA test — Existence | accuracy: 63.39 -> 69.8 | Helped |
| Add oracle object detections to HIMN instead of YOLO detections (ablation) | HIMN · IQA test — Counting | accuracy: 35.89 -> 32.2 | Hurt |
| Add oracle object detections to HIMN instead of YOLO detections (ablation) | HIMN · IQA test — Spatial Relationship | accuracy: 57.14 -> 65.6 | Helped |
| Add the oracle navigator with oracle object detection (ablation) | HIMN · IQA test — Counting | accuracy: 32.2 -> 45.35 | Helped |
| Add the oracle navigator with oracle object detection (ablation) | HIMN · IQA test — Existence | accuracy: 69.8 -> 73.03 | Helped |
| Add the oracle navigator with oracle object detection (ablation) | HIMN · IQA test — Spatial Relationship | accuracy: 65.6 -> 71.42 | Helped |
Closest papers by meaning
- Exemplar2VQA: A Scalable Exemplar-Driven Visual Question Answering Generation Framework via Multi-Agent Coding (NeurIPS 2026 · similarity 0.86)
- Multi-modal Situated Reasoning in 3D Scenes (NeurIPS 2024 · similarity 0.85)
- Divide-and-Conquer: Tree-structured Strategy with Answer Distribution Estimator for Goal-Oriented Visual Dialogue (AAAI 2025 · similarity 0.85)
- OpenEQA: Embodied Question Answering in the Era of Foundation Models (CVPR 2024 · similarity 0.85)
- ConVQG: Contrastive Visual Question Generation with Multimodal Guidance (AAAI 2024 · similarity 0.85)
- Question-Aware Gaussian Experts for Audio-Visual Question Answering (CVPR 2025 · similarity 0.84)
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.