EvalVerse: Pipeline-Aware and Expert-Calibrated Benchmarking for Professional Cinematic Video Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911707875508224 |
|---|---|
| author | Yang, Songlin Zhong, Haobin Zhang, Ruilin Zhao, Xiaotong Li, Shuai Zheng, Kai Yang, Xuyi Wang, Zhe Tang, Zhenchen Li, Yang Gu, Bohai Peng, Zhengwei Huang, Yidan Luo, Mengzhou Bo, Yihang Feng, Dalu Zhang, Yujia Ma, Juntao Wang, Ruiqi Zhang, Lvmin Guo, Yuwei Guan, Frank Agrawala, Maneesh Fu, Hongbo Zhao, Alan Rao, Anyi |
| author_facet | Yang, Songlin Zhong, Haobin Zhang, Ruilin Zhao, Xiaotong Li, Shuai Zheng, Kai Yang, Xuyi Wang, Zhe Tang, Zhenchen Li, Yang Gu, Bohai Peng, Zhengwei Huang, Yidan Luo, Mengzhou Bo, Yihang Feng, Dalu Zhang, Yujia Ma, Juntao Wang, Ruiqi Zhang, Lvmin Guo, Yuwei Guan, Frank Agrawala, Maneesh Fu, Hongbo Zhao, Alan Rao, Anyi |
| contents | The rapid evolution of generative video foundation models has propelled the field toward professional-grade cinematic synthesis. To achieve such demanding quality, the community transitions towards Reinforcement Learning (RL) and agentic workflows. However, reliable evaluation has emerged as a critical bottleneck. Existing benchmarks predominantly evaluate ''whether it is right'' (basic prompt-following) while fundamentally neglecting ''whether it is good'' (cinematic quality, acting, and aesthetics). Furthermore, current automated metrics lack the domain-specific rigor required to provide trustworthy signals, creating a severe credibility gap between human aesthetic perception and machine scoring. To bridge this gap, we introduce EvalVerse, a comprehensive, pipeline-aware, and expert-calibrated evaluation framework. We treat video generation assessment not merely as an engineering task, but as a core scientific problem: the systematic digitization of subjective cinematic expertise. First, we organize domain knowledge into an evaluation taxonomy aligned with the professional filmmaking workflow (pre-production, production, and post-production). Second, we distill human expert judgments into a curated dataset with large-scale human annotations. Third, we inject this knowledge into Vision-Language Models (VLMs) through an expert-calibrated fine-tuning strategy, enabling the VLM to perform explicit Chain-of-Thought reasoning. Compared to previous works, EvalVerse not only retains compatibility with foundational ''rightness'' metrics, but also significantly expands the criteria to ''goodness'' and broaden the task coverage to complex multi-shot sequencing and audio-visual integration. Consequently, by providing granular diagnostic signals, EvalVerse transcends a static leaderboard and establishes a fundamental infrastructure for future work, such as reward models and evaluator agent. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_23271 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | EvalVerse: Pipeline-Aware and Expert-Calibrated Benchmarking for Professional Cinematic Video Generation Yang, Songlin Zhong, Haobin Zhang, Ruilin Zhao, Xiaotong Li, Shuai Zheng, Kai Yang, Xuyi Wang, Zhe Tang, Zhenchen Li, Yang Gu, Bohai Peng, Zhengwei Huang, Yidan Luo, Mengzhou Bo, Yihang Feng, Dalu Zhang, Yujia Ma, Juntao Wang, Ruiqi Zhang, Lvmin Guo, Yuwei Guan, Frank Agrawala, Maneesh Fu, Hongbo Zhao, Alan Rao, Anyi Computer Vision and Pattern Recognition Artificial Intelligence The rapid evolution of generative video foundation models has propelled the field toward professional-grade cinematic synthesis. To achieve such demanding quality, the community transitions towards Reinforcement Learning (RL) and agentic workflows. However, reliable evaluation has emerged as a critical bottleneck. Existing benchmarks predominantly evaluate ''whether it is right'' (basic prompt-following) while fundamentally neglecting ''whether it is good'' (cinematic quality, acting, and aesthetics). Furthermore, current automated metrics lack the domain-specific rigor required to provide trustworthy signals, creating a severe credibility gap between human aesthetic perception and machine scoring. To bridge this gap, we introduce EvalVerse, a comprehensive, pipeline-aware, and expert-calibrated evaluation framework. We treat video generation assessment not merely as an engineering task, but as a core scientific problem: the systematic digitization of subjective cinematic expertise. First, we organize domain knowledge into an evaluation taxonomy aligned with the professional filmmaking workflow (pre-production, production, and post-production). Second, we distill human expert judgments into a curated dataset with large-scale human annotations. Third, we inject this knowledge into Vision-Language Models (VLMs) through an expert-calibrated fine-tuning strategy, enabling the VLM to perform explicit Chain-of-Thought reasoning. Compared to previous works, EvalVerse not only retains compatibility with foundational ''rightness'' metrics, but also significantly expands the criteria to ''goodness'' and broaden the task coverage to complex multi-shot sequencing and audio-visual integration. Consequently, by providing granular diagnostic signals, EvalVerse transcends a static leaderboard and establishes a fundamental infrastructure for future work, such as reward models and evaluator agent. |
| title | EvalVerse: Pipeline-Aware and Expert-Calibrated Benchmarking for Professional Cinematic Video Generation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2605.23271 |