Research map / Vision-language and multimodal / Visual question answering
HybridPrompt: Bridging Language Models and Human Priors in Prompt Tuning for Visual Question Answering
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 |
|---|---|---|---|
| Add answer filling verification loss (ablation) | HybridPrompt · VQAv2 test-dev | accuracy: 75.42 -> 76.12 | Helped |
| Add answer filling verification loss (ablation) | HybridPrompt · VQAv2 test-std | accuracy: 75.68 -> 76.30 | Helped |
| Add fixed-length trainable soft prompts (ablation) | HybridPrompt · VQAv2 test-dev | accuracy: 75.06 -> 76.12 | Helped |
| Add fixed-length trainable soft prompts (ablation) | HybridPrompt · VQAv2 test-std | accuracy: 75.18 -> 76.30 | Helped |
| Add contrastive sample filling with negative templates (ablation) | HybridPrompt · VQAv2 test-dev | accuracy: 74.35 -> 76.12 | Helped |
| Add contrastive sample filling with negative templates (ablation) | HybridPrompt · VQAv2 test-std | accuracy: 74.49 -> 76.30 | Helped |
| Add adaptive topic classification (ablation) | HybridPrompt · VQAv2 test-dev | accuracy: 71.08 -> 76.12 | Helped |
| Add adaptive topic classification (ablation) | HybridPrompt · VQAv2 test-std | accuracy: 71.15 -> 76.30 | Helped |
| Add topic-, sample-, and attribute-aware weak prompt concatenation (ablation) | HybridPrompt · VQAv2 test-dev | accuracy: 68.21 -> 76.12 | Helped |
| Add topic-, sample-, and attribute-aware weak prompt concatenation (ablation) | HybridPrompt · VQAv2 test-std | accuracy: 68.35 -> 76.30 | Helped |
| Increase soft-prompt length to 16 tokens rather than using 4 tokens (ablation) | HybridPrompt · VQAv2 | accuracy | Hurt |
Closest papers by meaning
- Making the v in VQA Matter: Elevating the Role of Image Understanding in 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)
- Right this way: Can VLMs Guide Us to See More to Answer Questions? (NeurIPS 2024 · similarity 0.91)
- Rephrase, Augment, Reason: Visual Grounding of Questions for Vision-Language Models (ICLR 2024 · similarity 0.90)
- FRAMES-VQA: Benchmarking Fine-Tuning Robustness across Multi-Modal Shifts in Visual Question Answering (CVPR 2025 · similarity 0.90)
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.