Research map / Vision-language and multimodal / Human and embodied modeling
Multi-Modal Dynamic Graph Transformer for Visual Grounding
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 |
|---|---|---|---|
| Use progressive graph-based region optimization instead of one-step region ranking | M-DGT · Flickr30k Entities test | Acc@0.5: 76.74 -> 79.97 | Helped |
| Use progressive graph-based region optimization instead of one-step region ranking | M-DGT · RefCOCO | accuracy: 82.68 -> 85.37 | Helped |
| Use progressive graph-based region optimization instead of one-step region ranking | M-DGT · RefCOCO+ | accuracy: 67.70 -> 70.02 | Helped |
| Use progressive graph-based region optimization instead of one-step region ranking | M-DGT · RefCOCOg | accuracy: 75.73 -> 79.21 | Helped |
| Use more grounding iterations, from 3 to 7 (ablation) | M-DGT · Flickr30k Entities | Acc@0.5: 69.815 -> 79.317 | Helped |
| Increase grounding iterations from 7 to 9 (ablation) | M-DGT · Flickr30k Entities | Acc@0.5: 79.317 -> 79.315 | No clear effect |
| Initialize the graph using Faster R-CNN proposals rather than simple image-covering boxes (ablation) | M-DGT · Flickr30k Entities | Acc@0.5: 79.317 -> 74.901 | Hurt |
| Initialize the graph using Bottom-Up Attention proposals rather than simple image-covering boxes (ablation) | M-DGT · Flickr30k Entities | Acc@0.5: 79.317 -> 76.162 | Hurt |
Closest papers by meaning
- SimVG: A Simple Framework for Visual Grounding with Decoupled Multi-modal Fusion (NeurIPS 2024 · similarity 0.89)
- Parallel Vertex Diffusion for Unified Visual Grounding (AAAI 2024 · similarity 0.88)
- GeM-VG: Towards Generalized Multi-image Visual Grounding with Multimodal Large Language Models (AAAI 2026 · similarity 0.87)
- Connecting the Dots: Training-Free Visual Grounding via Agentic Reasoning (AAAI 2026 · similarity 0.86)
- Multi-task Visual Grounding with Coarse-to-Fine Consistency Constraints (AAAI 2025 · similarity 0.86)
- AugRefer: Advancing 3D Visual Grounding via Cross-Modal Augmentation and Spatial Relation-based Referring (AAAI 2025 · similarity 0.86)
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.