Research map / Vision-language and multimodal / Cross-modal learning
Synergistic Prompting for Robust Visual Recognition with Missing Modalities
Zhihui Zhang, Luanyuan Dai, Qika Lin, Yunfeng Diao, Guangyin Jin, Yufei Guo, Jing Zhang, Xiaoshuai Hao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 synergistic prompts (dynamic plus static) to frozen-encoder classifier (ablation) | SyP · Hateful Memes 50% both missing | AUROC: 57.35 -> 68.16 | Improved |
| add synergistic prompts (dynamic plus static) to frozen-encoder classifier (ablation) | SyP · Food101 50% both missing | accuracy: 71.59 -> 86.17 | Improved |
| add synergistic prompts (dynamic plus static) to frozen-encoder classifier (ablation) | SyP · MM-IMDb 50% both missing | F1-Macro: 44.63 -> 54.72 | Improved |
| add static prompt to dynamic prompt (ablation) | SyP · Food101 50% both missing | accuracy: 82.90 -> 86.17 | Improved |
| add dynamic prompt to static prompt (ablation) | SyP · Hateful Memes 50% both missing | AUROC: 65.62 -> 68.16 | Improved |
| add dynamic adapter to base prompt (ablation) | SyP · Hateful Memes 50% both missing | AUROC: 64.27 -> 66.37 | Improved |
| add dynamic adapter to base prompt (ablation) | SyP · MM-IMDb 50% both missing | F1-Macro: 48.95 -> 51.21 | Improved |
| add dynamic adapter to synergistic prompts (ablation) | SyP · Food101 50% both missing | accuracy: 84.85 -> 86.17 | Improved |
| add dynamic adapter to synergistic prompts (ablation) | SyP · MM-IMDb 50% both missing | F1-Macro: 51.90 -> 54.72 | Improved |
| add dynamic adapter to synergistic prompts (ablation) | SyP · Hateful Memes 50% both missing | AUROC: 66.19 -> 68.16 | Improved |
| use SyP instead of DCP at 90% both-missing | SyP · Hateful Memes test 90% both missing | AUROC: 66.78 -> 68.93 | Improved |
| use SyP instead of DCP at 90% both-missing | SyP · MM-IMDb test 90% both missing | F1-Macro: 48.04 -> 49.63 | Improved |
Closest papers by meaning
- Deep Correlated Prompting for Visual Recognition with Missing Modalities (NeurIPS 2024 · similarity 0.92)
- AutoVP: An Automated Visual Prompting Framework and Benchmark (ICLR 2024 · similarity 0.89)
- Enhancing Visual Prompting through Expanded Transformation Space and Overfitting Mitigation (NeurIPS 2025 · similarity 0.88)
- SA²VP: Spatially Aligned-and-Adapted Visual Prompt (AAAI 2024 · similarity 0.88)
- LaViP: Language-Grounded Visual Prompting (AAAI 2024 · similarity 0.88)
- LoR-VP: Low-Rank Visual Prompting for Efficient Vision Model Adaptation (ICLR 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.
