Research map / Vision-language and multimodal / Visual question answering
Information Maximizing Visual Question Generation
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 |
|---|---|---|---|
| maximize mutual information with the expected answer using answer reconstruction (ablation) | Information Maximizing Visual Question Generator · VQA test, z-space | METEOR: 12.78 -> 18.78 | Helped |
| maximize mutual information with the expected answer using answer reconstruction (ablation) | Information Maximizing Visual Question Generator · VQA test, t-space | METEOR: 12.11 -> 18.35 | Helped |
| maximize mutual information with the expected answer using answer reconstruction (ablation) | Information Maximizing Visual Question Generator · VQA test, t-space | category accuracy: 68.23 -> 99.02 | Helped |
| add the category-conditioned second latent space as a regularizer (ablation) | Information Maximizing Visual Question Generator · VQA test, z-space | METEOR: 20.58 -> 18.78 | Hurt |
| add the category-conditioned second latent space as a regularizer (ablation) | Information Maximizing Visual Question Generator · VQA test, z-space | category accuracy: 61.04 -> 91.02 | Helped |
| remove answer and image reconstruction losses and the second latent space (ablation) | Information Maximizing Visual Question Generator · VQA test, z-space | category accuracy: 60.00 -> 91.02 | Hurt |
| use the full model rather than V-IC2Q | Information Maximizing Visual Question Generator · VQA test, overall | generation inventiveness: 38.32 -> 52.11 | Helped |
| use the full model rather than V-IC2Q | Information Maximizing Visual Question Generator · VQA test, overall | generation strength: 12.97 -> 26.06 | Helped |
Closest papers by meaning
- ConVQG: Contrastive Visual Question Generation with Multimodal Guidance (AAAI 2024 · similarity 0.90)
- An Empirical Evaluation of Visual Question Answering for Novel Objects (CVPR 2017 · similarity 0.90)
- VQ-VA World: Towards High-Quality Visual Question-Visual Answering (CVPR 2026 · similarity 0.89)
- Davidsonian Scene Graph: Improving Reliability in Fine-grained Evaluation for Text-to-Image Generation (ICLR 2024 · similarity 0.89)
- Divide-and-Conquer: Tree-structured Strategy with Answer Distribution Estimator for Goal-Oriented Visual Dialogue (AAAI 2025 · similarity 0.88)
- Retrieval-Augmented Visual Question Answering via Built-in Autoregressive Search Engines (AAAI 2025 · similarity 0.87)
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.