Research map / Vision-language and multimodal / Visual question answering
GQA: A New Dataset for Real-World Visual Reasoning and Compositional Question Answering
Check 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 |
|---|---|---|---|
| Add image features to the question-only LSTM | LSTM · GQA test | accuracy: 41.07 -> 46.55 | Helped |
| Use MAC instead of CNN+LSTM | CNN+LSTM · GQA test | accuracy: 46.55 -> 54.06 | Helped |
| Fine-tune the GQA-trained MAC model on VQA | MAC trained on GQA · VQA | accuracy: 52.1% -> 60.5% | Helped |
| Fine-tune the VQA-trained MAC model on GQA | MAC trained on VQA · GQA | accuracy: 39.8% -> 46.5% | Helped |
| Use object-informed visual features instead of spatial features | Attention-based models · GQA test | grounding: 43% | Helped |
| Smooth answer distributions using the question programs | GQA dataset construction · GQA answer distribution | entropy: 72% | Helped |
Closest papers by meaning
- OK-VQA: A Visual Question Answering Benchmark Requiring External Knowledge (CVPR 2019 · similarity 0.92)
- SQuINTing at VQA Models: Introspecting VQA Models With Sub-Questions (CVPR 2020 · similarity 0.92)
- LoRA: A Logical Reasoning Augmented Dataset for Visual Question Answering (NeurIPS 2023 · similarity 0.92)
- When Open-Vocabulary Visual Question Answering Meets Causal Adapter: Benchmark and Approach (AAAI 2025 · similarity 0.92)
- Core-to-Global Reasoning for Compositional Visual Question Answering (AAAI 2025 · similarity 0.91)
- VTQA: Visual Text Question Answering via Entity Alignment and Cross-Media Reasoning (CVPR 2024 · 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.