Research map / Vision-language and multimodal / Human and embodied modeling
Connective Cognition Network for Directional Visual Commonsense Reasoning
Aming Wu, Linchao Zhu, Yahong Han, Yi Yang
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 conditional centers and GCN to NetVLAD (GraphVLAD) (ablation) | CCN · VCR Q->AR val | accuracy (%): 45.6 -> 47.7 | Improved |
| add conditional centers (affine-transformed) to GraphVLAD (ablation) | CCN · VCR Q->AR val | accuracy (%): 46.5 -> 47.7 | Improved |
| add GCN over NetVLAD outputs (ablation) | CCN · VCR Q->AR val | accuracy (%): 46.6 -> 47.7 | Improved |
| add reasoning module (ablation) | CCN · VCR Q->AR val | accuracy (%): 45.3 -> 47.7 | Improved |
| use directional GCN instead of undirected GCN for reasoning (ablation) | CCN · VCR Q->AR val | accuracy (%): 46.4 -> 47.7 | Improved |
| use directional GCN instead of LSTM reasoning (ablation) | CCN · VCR Q->AR val | accuracy (%): 43.9 -> 47.7 | Improved |
| LSTM reasoning instead of no reasoning (ablation) | CCN · VCR Q->AR val | accuracy (%): 45.3 -> 43.9 | Hurt |
| separate GCNs for query and response instead of one GCN on concatenation (ablation) | CCN · VCR Q->AR val | accuracy (%): 45.7 -> 47.7 | Improved |
| 16 centers instead of 32 in GraphVLAD (ablation) | CCN · VCR Q->AR val | accuracy (%): 47.7 -> 46.4 | Hurt |
| 48 centers instead of 32 in GraphVLAD (ablation) | CCN · VCR Q->AR val | accuracy (%): 47.7 -> 46.9 | Hurt |
Closest papers by meaning
- Toward Cognitive Supersensing in Multimodal Large Language Models (NeurIPS 2026 · similarity 0.85)
- VGR: Visual Grounded Reasoning (ICLR 2026 · similarity 0.85)
- Visual Chain-of-Thought Prompting for Knowledge-Based Visual Reasoning (AAAI 2024 · similarity 0.85)
- Multimodal Commonsense Knowledge Distillation for Visual Question Answering (Student Abstract) (AAAI 2025 · similarity 0.85)
- Monet: Reasoning in Latent Visual Space Beyond Image and Language (CVPR 2026 · similarity 0.84)
- Visual Graph Arena: Evaluating Visual Conceptualization of Vision and Multimodal Large Language Models (ICML 2025 · similarity 0.84)
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.
