Research map / Vision-language and multimodal / Visual question answering
Visual Question Answering with Question Representation Update (QRU)
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 |
|---|---|---|---|
| replace attention-weighted pooling with mean pooling (ablation) | QRU (1 reasoning layer) · COCO-QA | ACC.: 61.99 -> 58.15 without (about -3.84 with mean pooling) | Helped |
| replace attention-weighted pooling with max pooling (ablation) | QRU (1 reasoning layer) · COCO-QA | ACC.: 61.99 -> 59.37 | Helped |
| remove whole-image global region from image features (ablation) | QRU (1 reasoning layer) · COCO-QA | ACC.: 61.99 -> 60.87 (drop of 1.12) | Helped |
| remove 8D spatial coordinate encoding of regions (ablation) | QRU (1 reasoning layer) · COCO-QA | ACC.: 61.99 -> 61.33 | Helped |
| increase reasoning layers from one to two | QRU · COCO-QA | ACC.: 61.99 -> 62.50 | Helped |
| increase reasoning layers from one to two | QRU · VQA test-dev Open-Ended | All: 59.26 -> 60.72 | Helped |
| use three or more reasoning layers instead of two | QRU · COCO-QA | ACC. | No clear effect |
| use object proposals (19 edge boxes plus whole image) instead of global attention over CNN feature map | QRU · VQA test-dev Open-Ended | All: Other type gain; QRU (2) 60.72 vs SAN 58.70 | Helped |
Closest papers by meaning
- Graph-Structured Representations for Visual Question Answering (CVPR 2017 · similarity 0.93)
- Self-Critical Reasoning for Robust Visual Question Answering (NeurIPS 2019 · similarity 0.93)
- Tips and Tricks for Visual Question Answering: Learnings From the 2017 Challenge (CVPR 2018 · similarity 0.93)
- Multi-Level Attention Networks for Visual Question Answering (CVPR 2017 · similarity 0.92)
- Don't Just Assume; Look and Answer: Overcoming Priors for Visual Question Answering (CVPR 2018 · similarity 0.92)
- When Open-Vocabulary Visual Question Answering Meets Causal Adapter: Benchmark and Approach (AAAI 2025 · similarity 0.91)
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.