Research map / Vision-language and multimodal / Multimodal LLMs
OmniDiff: A Comprehensive Benchmark for Fine-grained Image Difference Captioning
Yuan Liu, Saihui Hou, Saijie Hou, Jiabao Du, Shibei Meng, Yongzhen Huang
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 OmniDiff data to fine-tuning (ablation) | M3Diff · OmniDiff-Real (fine-tuned) | CIDEr: 1.9 -> 31.3 | Improved |
| add OmniDiff data to fine-tuning (ablation) | M3Diff · OmniDiff-Render (fine-tuned) | CIDEr: 1.1 -> 28.3 | Improved |
| add OmniDiff data to fine-tuning (ablation) | M3Diff · Image-Edit-Request | CIDEr: 132.8 -> 136.6 | Improved |
| add MDP module (ablation) | M3Diff · OmniDiff-Real (fine-tuned) | CIDEr: 35.3 -> 31.3 | Hurt |
| add MDP module (ablation) | M3Diff · OmniDiff-Real (fine-tuned) | BLEU-4: 12.2 -> 14.3 | Improved |
| add MDP module (ablation) | M3Diff · OmniDiff-Render (fine-tuned) | CIDEr: 25.4 -> 28.3 | Improved |
| add MDP module (ablation) | M3Diff · Image-Edit-Request | CIDEr: 135.2 -> 136.6 | Improved |
| add MDP module (ablation) | Qwen2.5-VL-7B SFT · OmniDiff-Real (fine-tuned) | CIDEr: 35.3 -> 36.7 | Improved |
| add MDP module (ablation) | InternVL3-8B SFT · OmniDiff-Real (fine-tuned) | CIDEr: 34.6 -> 35.7 | Improved |
| add MDP module (ablation) | Qwen2.5-VL-7B SFT · Image-Edit-Request | CIDEr: 131.2 -> 132.5 | Improved |
| add MDP module (ablation) | InternVL3-8B SFT · Image-Edit-Request | CIDEr: 128.1 -> 130.5 | Improved |
| add OmniDiff data to fine-tuning (GPT-4o LLM score) (ablation) | M3Diff · OmniDiff-Real (fine-tuned) | LLM score: 12.3 -> 37.3 | Improved |
Closest papers by meaning
- Describing Differences in Image Sets with Natural Language (CVPR 2024 · similarity 0.91)
- Img-Diff: Contrastive Data Synthesis for Multimodal Large Language Models (CVPR 2025 · similarity 0.89)
- VisMin: Visual Minimal-Change Understanding (NeurIPS 2024 · similarity 0.89)
- Region-aware Difference Distilling with Attribute-guided Contrastive Regularization for Change Captioning (AAAI 2025 · similarity 0.88)
- VIXEN: Visual Text Comparison Network for Image Difference Captioning (AAAI 2024 · similarity 0.88)
- Semantic and Expressive Variations in Image Captions Across Languages (CVPR 2025 · similarity 0.87)
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.
