Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
video
video
8.31
59.6
label
class label
8 classes
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
0cat_movie
1holocine
1holocine
1holocine
1holocine
1holocine
1holocine
1holocine
1holocine
End of preview. Expand in Data Studio

CaT-MBench-Videos

Project Page GitHub Code HF Collection CaT-MBench Weights License

All 736 AI-generated evaluation videos (92 test stories × 8 methods, 13.65 GiB, seed 42) reported in Table 1 of CaT-Movie, paired with JinPLU/CaT-MBench for side-by-side visual inspection and end-to-end re-scoring.

Included Methods (92 Stories Each)

Each folder stores 92 MP4 files named <story_id>.mp4 matching stories.json in JinPLU/CaT-MBench, indexed by manifest.json:

Directory Method Video Base
cat_movie/ CaT-Movie (Ours) Wan2.2-TI2V-5B
wan22_ti2v_5b_official_per_shot/ Wan2.2, per shot Wan2.2-TI2V-5B
wan21_t2v_1.3b_official_per_shot/ Wan2.1, per shot Wan2.1-T2V-1.3B
holocine/ HoloCine Wan2.2-T2V-A14B
multishotmaster/ MultiShotMaster Wan2.1-T2V-1.3B
storymem/ StoryMem Wan2.2-I2V-A14B
storydiffusion/ StoryDiffusion Wan2.2-I2V-A14B
ic_lora/ IC-LoRA Wan2.2-I2V-A14B

Download

# Full 736-video comparison set (13.65 GiB)
hf download JinPLU/CaT-MBench-Videos --repo-type dataset --local-dir data/cat_mbench_videos

# Or download a single method folder (e.g., CaT-Movie)
hf download JinPLU/CaT-MBench-Videos --repo-type dataset \
  --include "cat_movie/*" --local-dir data/cat_mbench_videos
CaT-Movie Release Ecosystem (GitHub & 8 HF Repositories)
Repository Tier Contents
JinPLu/CaT-Movie (GitHub) Core Planner, training, inference, curation & CaT-MBench code (Apache-2.0)
JinPLU/CaT-Movie Core Renderer weights (step-3000.safetensors)
JinPLU/CaT-Movie-Data Core 75,461 blocks, Qwen3.7-Flash schemas, grounding, 96,000 train windows & SAM 3.1 weights
JinPLU/CaT-Movie-T5-Cache Core Precomputed umT5-XXL BF16 caches (332,516 unique prompts)
JinPLU/CaT-Movie-VAE-Caches Core Precomputed VAE target latents (Wan2.2 704p + Wan2.1 480p)
JinPLU/CaT-Movie-Source-Videos Supporting Complete 18,173 block-yielding source MP4 clips
JinPLU/CaT-MBench Core 122 stories, cut-gated GPT-6.1-sol VQA bank, Gemini 3.8 Flash verdicts & 557 human labels
JinPLU/CaT-MBench-Videos (this repo) Supporting 736 generated test videos (92 test stories × 8 methods)
JinPLU/CaT-MBench-Author-Labels Archive Early Nov 6 single-author calibration archive (superseded by CaT-MBench/human/)

Attribution, Licensing & Content Notice

  • Synthetic Media Disclosure: All 736 MP4 files are 100% synthetic AI-generated videos produced purely from text/storyboard prompts (zero ground-truth footage redistributed).
  • License Terms: Packaged under CC BY-NC-SA 4.0 strictly for non-commercial research. Baseline generations inherit upstream model terms (Wan2.1 / Wan2.2 / StoryMem / HoloCine / MultiShotMaster / IC-LoRA: Apache-2.0; StoryDiffusion: see upstream repository; IC-LoRA outputs underwent content screening before release).
  • Content Screening: Story prompts derive from screened cinematic dialogue/action scenes; outputs may still reflect artifacts inherited from upstream foundation models.

Citation

@misc{catmovie2026,
  title         = {CaT-Movie: Balancing Global Planning and Local Rendering for Long Multi-Shot Video Generation},
  author        = {Jinpeng Lu and Haoyuan Shi and Yuanshen Guan and Jiyuan Xia and Bin Wu and Yinda Chen and Zhiwei Xiong},
  year          = {2026},
  url           = {https://jinplu.github.io/CaT-Movie/}
}
Downloads last month
-

Collection including JinPLU/CaT-MBench-Videos