Research map / Vision-language and multimodal / Human and embodied modeling
Image Enhanced Event Detection in News Articles
Meihan Tong, Shuai Wang, Yixin Cao, Bin Xu, Juanzi Li, Lei Hou, Tat-Seng Chua
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 images with dual recurrent multimodal model (DRMM) on top of BERT (ablation) | BERT · ACE2005 ED | F1: 75.1 -> 76.3 | Improved |
| add image modality (ablation) | CNN encoder · ACE2005 ED | F1: 59.9 -> 63.3 | Improved |
| add image modality (ablation) | LSTM encoder · ACE2005 ED | F1: 60.2 -> 64.8 | Improved |
| use alternating dual attention instead of concatenation fusion (ablation) | DRMM · ACE2005 ED | F1: 69.2 -> 76.3 | Improved |
| use alternating dual attention instead of modality attention (ablation) | DRMM · ACE2005 ED | F1: 73.8 -> 76.3 | Improved |
| use alternating dual attention instead of co-attention (ablation) | DRMM · ACE2005 ED | F1: 74.6 -> 76.3 | Improved |
| add images to masked language model (ablation) | language model · ACE2005 MLM | F1: 75.8 -> 79.4 | Improved |
| add image modality (BERT+image) for zero-shot and few-shot triggers (ablation) | BERT · ACE2005 ED, trigger frequency buckets | Precision: +3.9% zero-shot, +2.7% few-shot | Improved |
| add image modality (BERT+image) for high-frequency triggers (ablation) | BERT · ACE2005 ED, high-frequency triggers | Precision: -0.7% | No clear effect |
| DRMM (BERT+images) vs text-only AD-DMBERT | event detection · ACE2005 ED | F1: 75.1 -> 76.3 | Improved |
Closest papers by meaning
- Cross-modal Multi-task Learning for Multimedia Event Extraction (AAAI 2025 · similarity 0.85)
- Beyond Grounding: Extracting Fine-Grained Event Hierarchies across Modalities (AAAI 2024 · similarity 0.85)
- EventGPT: Event Stream Understanding with Multimodal Large Language Models (CVPR 2025 · similarity 0.84)
- Event-Equalized Dense Video Captioning (CVPR 2025 · similarity 0.83)
- Dependency Structure-Enhanced Graph Attention Networks for Event Detection (AAAI 2024 · similarity 0.83)
- OpenESS: Event-based Semantic Scene Understanding with Open Vocabularies (CVPR 2024 · similarity 0.83)
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.
