Research map / Video / Video-language understanding
A Benchmark and Multi-Agent System for Instruction-driven Cinematic Video Compilation
Peixuan Zhang, Chang Zhou, Ziyuan Zhang, Hualuo Liu, Chunjie Zhang, Jingqi Liu, Xiaohui Zhou, Xi Chen, Shuchen Weng, Si Li, Boxin Shi
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 dialogue character matching (ablation) | CineAgents · CineBench | SVC: 0.148 -> 0.154 | Improved |
| add dialogue character matching (ablation) | CineAgents · CineBench | F1-score: 61.17 -> 64.13 | Improved |
| add hierarchical narrative memory (ablation) | CineAgents · CineBench | F1-score: 58.04 -> 64.13 | Improved |
| add hierarchical narrative memory (ablation) | CineAgents · CineBench | Precision: 49.12 -> 62.43 | Improved |
| add hierarchical narrative memory (ablation) | CineAgents · CineBench | ARR: 72.96% -> 87.23% | Improved |
| add iterative narrative planning instead of retrieve-and-rank (ablation) | CineAgents · CineBench | ARR: 78.81% -> 87.23% | Improved |
| add iterative narrative planning instead of retrieve-and-rank (ablation) | CineAgents · CineBench | F1-score: 59.52 -> 64.13 | Improved |
| add iterative narrative planning instead of retrieve-and-rank (ablation) | CineAgents · CineBench | TCS: 49.27% -> 52.09% | Improved |
| add dialogue character matching (ablation) | CineAgents · CineBench | ARR: 85.34% -> 87.23% | Improved |
| add dialogue character matching (ablation) | CineAgents · CineBench | ESR: 89.91% -> 92.76% | Improved |
| use CineAgents design-and-compose instead of plain Gemini retrieve-and-rank | Gemini-2.5-pro · CineBench | F1-score: 41.59 -> 64.13 | Improved |
| use CineAgents instead of MetaGPT multi-agent framework | MetaGPT · CineBench | Precision: 43.71 -> 62.43 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- VideoAgent — VideoAgent: A Memory-augmented Multimodal Agent for Video Understanding (ECCV 2024) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- CineTechBench: A Benchmark for Cinematographic Technique Understanding and Generation (NeurIPS 2025 · similarity 0.86)
- MovieDreamer: Hierarchical Generation for Coherent Long Visual Sequences (ICLR 2025 · similarity 0.84)
- Stable Cinemetrics : Structured Taxonomy and Evaluation for Professional Video Generation (NeurIPS 2025 · similarity 0.84)
- CineOrchestra: Unified Entity-Centric Conditioning for Cinematic Video Generation (NeurIPS 2026 · similarity 0.84)
- LogiStory: A Logic-Aware Framework for Multi-Image Story Visualization (ICLR 2026 · similarity 0.83)
- MovieBench: A Hierarchical Movie Level Dataset for Long Video Generation (CVPR 2025 · similarity 0.82)
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.
