Research map / Vision-language and multimodal / Human and embodied modeling
Describing and Localizing Multiple Changes With Transformers
Yue Qiu, Shintaro Yamamoto, Kodai Nakashima, Ryota Suzuki, Kenji Iwata, Hirokatsu Kataoka, Yutaka Satoh
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 |
|---|---|---|---|
| replace the DUDA decoder with a transformer decoder (ablation) | DUDA · CLEVR-Multi-Change | BLEU-4: 76.1 -> 79.1 | Improved |
| replace the M-VAM decoder with a transformer decoder (ablation) | M-VAM · CLEVR-Multi-Change | BLEU-4: 62.9 -> 65.8 | Improved |
| concatenate before/after features over feature dimension rather than over patches (ablation) | MCCFormers-D · CLEVR-Multi-Change | BLEU-4: 80.1 -> 82.3 | Improved |
| concatenate before/after features over feature dimension rather than over patches (ablation) | MCCFormers-S · CLEVR-Multi-Change | BLEU-4: 80.6 -> 83.3 | Improved |
| use MCCFormers-S (single encoder) instead of D (dual) encoder | MCCFormers · CLEVR-Change | BLEU-4: 52.4 -> 57.4 | Improved |
| use MCCFormers-D instead of S encoder | MCCFormers · Spot-the-Diff (multiple change) | CIDEr: 18.2 -> 28.8 | Improved |
| use MCCFormers-D over DUDA on multi-change instances | MCCFormers-D · CLEVR-Multi-Change (4 changes) | BLEU-4: 70.9 -> 80.6 | Improved |
| use MCCFormers-D over DUDA for sentence number | MCCFormers-D · CLEVR-Multi-Change | sentence number accuracy: 83.3 -> 92.5 | Improved |
| use MCCFormers-D over DUDA on Spot-the-Diff multi-change BLEU-4 | MCCFormers-D · Spot-the-Diff (multiple change) | BLEU-4: 5.4 -> 6.2 | No clear effect |
| use MCCFormers-D instead of M-VAM + RAF on Spot-the-Diff single change | MCCFormers-D · Spot-the-Diff (single change) | BLEU-4: 11.1 -> 10.0 | No clear effect |
| use 2 encoder/decoder layers instead of 1 layer (1 head) (ablation) | MCCFormers-D · CLEVR-Multi-Change | BLEU-4: 59.0 -> 81.4 | Improved |
| use 4 layers instead of 2 layers (4 heads) (ablation) | MCCFormers-D · CLEVR-Multi-Change | BLEU-4: 82.3 -> 64.8 | Hurt |
Closest papers by meaning
- Region-aware Difference Distilling with Attribute-guided Contrastive Regularization for Change Captioning (AAAI 2025 · similarity 0.90)
- Revisiting Change Captioning from Self-supervised Global-Part Alignment (AAAI 2025 · similarity 0.88)
- Change3D: Revisiting Change Detection and Captioning from A Video Modeling Perspective (CVPR 2025 · similarity 0.87)
- VisMin: Visual Minimal-Change Understanding (NeurIPS 2024 · similarity 0.86)
- DECIDER: Difference-aware Contrastive Diffusion Model with Adversarial Perturbations for Image Change Captioning (AAAI 2025 · similarity 0.85)
- Describing Differences in Image Sets with Natural Language (CVPR 2024 · 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.
