Research map / Graph learning / Knowledge and temporal graphs
G3raphGround: Graph-Based Language Grounding
Mohit Bajaj, Lanjun Wang, Leonid Sigal
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 the phrase graph (ablation) | G3RAPHGROUND · Flickr30k Entities test | phrase grounding accuracy (%): 60.82 -> 63.87 | Improved |
| add the visual graph (ablation) | G3RAPHGROUND · Flickr30k Entities test | phrase grounding accuracy (%): 62.23 -> 63.87 | Improved |
| add the fusion graph (ablation) | G3RAPHGROUND · Flickr30k Entities test | phrase grounding accuracy (%): 59.13 -> 63.87 | Improved |
| add the fusion graph (ablation) | G3RAPHGROUND · ReferIt Game test | phrase grounding accuracy (%): 36.54 -> 41.79 | Improved |
| add the visual graph and fusion graph together (ablation) | G3RAPHGROUND · Flickr30k Entities test | phrase grounding accuracy (%): 56.32 -> 63.87 | Improved |
| add the full image embedding as context (ablation) | G3RAPHGROUND · Flickr30k Entities test | phrase grounding accuracy (%): 62.32 -> 63.87 | Improved |
| add the full caption embedding as context (ablation) | G3RAPHGROUND · Flickr30k Entities test | phrase grounding accuracy (%): 62.73 -> 63.87 | Improved |
| add the full caption embedding as context (ablation) | G3RAPHGROUND · ReferIt Game test | phrase grounding accuracy (%): 41.79 -> 41.79 | No clear effect |
| add the full image embedding as context (ablation) | G3RAPHGROUND · ReferIt Game test | phrase grounding accuracy (%): 40.92 -> 41.79 | Improved |
| use per-phrase conditional fusion graphs instead of one big fusion graph (ablation) | GGFusionBase · Flickr30k Entities test | phrase grounding accuracy (%): 60.41 -> 63.87 | Improved |
| add count prediction branch replacing threshold post-processing (ablation) | G3RAPHGROUND · Flickr30k Entities test | phrase grounding accuracy (%): 63.87 -> 66.93 | Improved |
| add count prediction branch replacing threshold post-processing (ablation) | G3RAPHGROUND · ReferIt Game test | phrase grounding accuracy (%): 41.79 -> 44.91 | Improved |
Closest papers by meaning
- Learning Cross-Modal Context Graph for Visual Grounding (AAAI 2020 · similarity 0.91)
- GeM-VG: Towards Generalized Multi-image Visual Grounding with Multimodal Large Language Models (AAAI 2026 · similarity 0.88)
- Connecting the Dots: Training-Free Visual Grounding via Agentic Reasoning (AAAI 2026 · similarity 0.87)
- AugRefer: Advancing 3D Visual Grounding via Cross-Modal Augmentation and Spatial Relation-based Referring (AAAI 2025 · similarity 0.87)
- Chain of Semantics Programming in 3D Gaussian Splatting Representation for 3D Vision Grounding (CVPR 2025 · similarity 0.86)
- GLaMM: Pixel Grounding Large Multimodal Model (CVPR 2024 · 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.
