Research map / Video / Video generation
CI-VID: A Coherent Interleaved Text-Video Dataset
Yiming Ju, Jijin Hu, Zhengxiong Luo, Haoge Deng, hanyu Zhao, Li Du, Wenbo Xiao, Chengwei Wu, Donglin Hao, Xinlong Wang, Tengfei Pan
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 |
|---|---|---|---|
| fine-tune on CI-VID | NOVA-0.6B video model · Human evaluation (consistency) | win rate: 90.0% win, 6.5% tie, 3.6% loss | Improved |
| fine-tune on CI-VID | NOVA-0.6B video model · Human evaluation (narrativity) | win rate: 80.9% win, 15.0% tie, 4.1% loss | Improved |
| fine-tune on CI-VID | NOVA-0.6B video model · Human evaluation (correctness) | win rate: 78.3% win, 9.8% tie, 11.9% loss | Improved |
| fine-tune on CI-VID | NOVA-0.6B video model · VLM evaluation (Stylistic Consistency) | VLM score: 2.93 -> 3.83 | Improved |
| fine-tune on CI-VID | NOVA-0.6B video model · VLM evaluation (Entity Consistency) | VLM score: 2.84 -> 3.73 | Improved |
| fine-tune on CI-VID | NOVA-0.6B video model · VLM evaluation (Background Consistency) | VLM score: 2.80 -> 3.75 | Improved |
| fine-tune on CI-VID | NOVA-0.6B video model · VLM evaluation (Perspective Transition) | VLM score: 3.02 -> 3.81 | Improved |
| fine-tune on CI-VID | NOVA-0.6B video model · VLM evaluation (Text Prompt Alignment) | VLM score: 3.99 -> 4.07 | No clear effect |
| fine-tune on CI-VID | NOVA-0.6B video model · VLM evaluation (Visual Plausibility) | VLM score: 3.25 -> 3.62 | Improved |
| fine-tune on CI-VID | NOVA-0.6B video model · Similarity evaluation (Whole) | CLIP: 0.512 -> 0.670 | Improved |
| fine-tune on CI-VID | NOVA-0.6B video model · Similarity evaluation (Whole) | SSIM: 0.199 -> 0.272 | Improved |
| fine-tune on CI-VID | NOVA-0.6B video model · Similarity evaluation (Object) | CLIP: 0.601 -> 0.702 | Improved |
Closest papers by meaning
- OpenVid-1M: A Large-Scale High-Quality Dataset for Text-to-video Generation (ICLR 2025 · similarity 0.88)
- LVD-2M: A Long-take Video Dataset with Temporally Dense Captions (NeurIPS 2024 · similarity 0.88)
- CoMM: A Coherent Interleaved Image-Text Dataset for Multimodal Understanding and Generation (CVPR 2025 · similarity 0.87)
- T2V-CompBench: A Comprehensive Benchmark for Compositional Text-to-video Generation (CVPR 2025 · similarity 0.87)
- A Recipe for Scaling up Text-to-Video Generation with Text-free Videos (CVPR 2024 · similarity 0.87)
- Panda-70M: Captioning 70M Videos with Multiple Cross-Modality Teachers (CVPR 2024 · 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.
