Research map / Vision-language and multimodal / Human and embodied modeling
Aligning Visual Regions and Textual Concepts for Semantic-Grounded Image Representations
Read the paperCheck 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 |
|---|---|---|---|
| Add MIA-refined representations | Visual Regional Attention · MSCOCO image captioning | CIDEr: 111.2 -> 114.1 | Helped |
| Add MIA-refined representations | Transformer · MSCOCO image captioning | CIDEr: 126.3 -> 129.6 | Helped |
| Replace image features with MIA-refined representations | BAN · VQA v2.0 test-std | accuracy: 69.8 -> 70.3 | Helped |
| Use refined visual features only (ablation) | Visual Attention · MSCOCO image captioning | CIDEr: 103.0 -> 105.7 | Helped |
| Use refined textual concepts only (ablation) | Concept Attention · MSCOCO image captioning | CIDEr: 103.2 -> 106.5 | Helped |
| Replace mutual attention with self-attention (ablation) | Visual Attention · MSCOCO image captioning | CIDEr: 103.0 -> 96.6 | Hurt |
| Replace mutual attention with self-attention (ablation) | Concept Attention · MSCOCO image captioning | CIDEr: 103.2 -> 105.4 | Hurt |
| Make visual features guide textual concepts first (I_N -> T_N) (ablation) | Visual Attention · MSCOCO image captioning | CIDEr: 103.0 -> 99.0 | Hurt |
| Make visual features guide textual concepts first (I_N -> T_N) (ablation) | Concept Attention · MSCOCO image captioning | CIDEr: 103.2 -> 105.5 | Helped |
| Increase mutual-attention iterations beyond two (ablation) | MIA · MSCOCO image captioning | captioning metrics | Hurt |
Closest papers by meaning
- Your Large Vision-Language Model Only Needs A Few Attention Heads For Visual Grounding (CVPR 2025 · similarity 0.90)
- Remodeling Semantic Relationships in Vision-Language Fine-Tuning (AAAI 2026 · similarity 0.90)
- Look Around Before Locating: Considering Content and Structure Information for Visual Grounding (AAAI 2025 · similarity 0.90)
- Improved Visual Grounding through Self-Consistent Explanations (CVPR 2024 · similarity 0.89)
- Connecting the Dots: Training-Free Visual Grounding via Agentic Reasoning (AAAI 2026 · similarity 0.89)
- SimVG: A Simple Framework for Visual Grounding with Decoupled Multi-modal Fusion (NeurIPS 2024 · similarity 0.89)
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.