Research map / Vision-language and multimodal / Human and embodied modeling
Show, Edit and Tell: A Framework for Editing Image Captions
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 |
|---|---|---|---|
| Add MSE optimization after cross-entropy training (ablation) | DCNet · MS COCO Karpathy test | CIDEr-D: 117.1 -> 118.3 | Helped |
| Add MSE optimization after cross-entropy training (ablation) | DCNet · MS COCO Karpathy test | BLEU-4: 36.9 -> 37.2 | Helped |
| Add MSE optimization after cross-entropy training (ablation) | EditNet · MS COCO Karpathy test | CIDEr-D: 118.0 -> 118.5 | Helped |
| Add MSE optimization after cross-entropy training (ablation) | EditNet · MS COCO Karpathy test | BLEU-4: 38.0 -> 38.0 | No clear effect |
| Add visual features, context gating, and DCNet to Copy-LSTM alone (ablation) | ETN · MS COCO Karpathy test | CIDEr-D: 117.3 -> 1.200 | Helped |
| Add the context gate (ablation) | EditNet without visual features · MS COCO Karpathy test | CIDEr-D: 117.3 -> 117.5 | Helped |
| Replace non-differentiable indexing with soft attention on memory states (ablation) | SCMA · MS COCO Karpathy test | CIDEr-D: 1.200 -> 119.2 | Hurt |
| Use iterative caption editing with Copy-LSTM and SCMA | ETN · MS COCO Karpathy test, cross-entropy | CIDEr-D: 119.8 -> 1.200 | Helped |
| Use iterative caption editing with Copy-LSTM and SCMA | ETN · MS COCO Karpathy test, sequence-level optimization | CIDEr-D: 129.8 -> 128.9 | Hurt |
Closest papers by meaning
- Noise-Aware Image Captioning with Progressively Exploring Mismatched Words (AAAI 2024 · similarity 0.88)
- Incorporating Copying Mechanism in Image Captioning for Learning Novel Objects (CVPR 2017 · similarity 0.87)
- Patch Matters: Training-free Fine-grained Image Caption Enhancement via Local Perception (CVPR 2025 · similarity 0.86)
- CapRL: Stimulating Dense Image Caption Capabilities via Reinforcement Learning (ICLR 2026 · similarity 0.86)
- Painting with Words: Elevating Detailed Image Captioning with Benchmark and Alignment Learning (ICLR 2025 · similarity 0.86)
- An Item Is Worth a Prompt: Versatile Image Editing with Disentangled Control (AAAI 2025 · similarity 0.85)
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.