Research map / Graph learning / Knowledge and temporal graphs
Neighbourhood Watch: Referring Expression Comprehension via Language-Guided Graph Attention Networks
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 language-guided graph attention over objects and their relationships | LGRANs · RefCOCO testB, ground-truth regions | accuracy: 82.39 -> 84.0 | Helped |
| use language-guided graph attention over objects and their relationships | LGRANs · RefCOCO+ testA, ground-truth regions | accuracy: 68.76 -> 67.6 | Hurt |
| use language-guided graph attention over objects and their relationships | LGRANs · RefCOCO testB, automatically detected regions | accuracy: 67.44 -> 66.4 | Hurt |
| add object-relationship edge representations to isolated node features (ablation) | NodeRep · RefCOCOg val | accuracy: 67.1 -> 70.5 | Helped |
| add node attention to graph representations (ablation) | GraphRep · RefCOCOg val | accuracy: 70.5 -> 72.4 | Helped |
| add edge attention to graph representations (ablation) | GraphRep · RefCOCOg val | accuracy: 70.5 -> 73.9 | Helped |
| combine node attention and edge attention (ablation) | LGRANs · RefCOCOg val | accuracy: 72.4 -> 75.4 | Helped |
| add object-relationship edge representations to isolated node features (ablation) | NodeRep · RefCOCO val | accuracy: 77.6 -> 80.2 | Helped |
| add node attention to graph representations (ablation) | GraphRep · RefCOCO val | accuracy: 80.2 -> 81.4 | Helped |
| add edge attention to graph representations (ablation) | GraphRep · RefCOCO val | accuracy: 80.2 -> 81.9 | Helped |
| combine node attention and edge attention (ablation) | LGRANs · RefCOCO val | accuracy: 81.9 -> 82.0 | Helped |
| add node attention to graph representations (ablation) | GraphRep · RefCOCO+ testB | accuracy: 61.9 -> 64.2 | Helped |
Closest papers by meaning
- RefDetector: A Simple Yet Effective Matching-based Method for Referring Expression Comprehension (AAAI 2025 · similarity 0.90)
- Rethinking Two-Stage Referring Expression Comprehension: A Novel Grounding and Segmentation Method Modulated by Point (AAAI 2024 · similarity 0.89)
- ScanFormer: Referring Expression Comprehension by Iteratively Scanning (CVPR 2024 · similarity 0.89)
- Ref-Adv: Exploring MLLM Visual Reasoning in Referring Expression Tasks (ICLR 2026 · similarity 0.88)
- Boosting Weakly Supervised Referring Image Segmentation via Progressive Comprehension (NeurIPS 2024 · similarity 0.88)
- DViN: Dynamic Visual Routing Network for Weakly Supervised Referring Expression Comprehension (CVPR 2025 · similarity 0.88)
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.