UniVBench：面向视频基础模型的统一评测
UniVBench: Towards Unified Evaluation for Video Foundation Models
为视频基础模型提供首个统一评测基准，涵盖理解、生成、编辑和重建四项任务，并引入智能体评估系统标准化提示、指令解析和评分，便于研究者公平比较统一视频模型。
原文依据 we introduce UniVBench, a benchmark purpose-built for evaluating video foundation models across four core abilities
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论视频基础模型旨在将视频理解、生成、编辑和指令跟随集成到一个单一框架中，使其成为下一代多模态系统的核心方向。然而，现有的评测基准仍然碎片化且范围有限，因为它们各自针对单一任务，依赖任务特定的指标，并且通常使用短小或简单的视频片段。因此，它们无法捕捉这些模型设计所要提供的统一能力。为了弥补这一差距，我们推出了 UniVBench，一个专为评估视频基础模型在四个核心能力上表现而构建的基准：视频理解、视频生成、视频编辑，以及一个新提出的任务——视频重建，该任务评估模型重现其遇到过的视频内容的忠实程度。我们的基准通过纳入 200 个高质量、多样且多镜头的视频，每个视频都配有详细的描述、多格式编辑指令和参考图像，大幅扩展了评测的复杂性。所有视频完全由人类创作并经过仔细验证，提供了比先前基准更丰富的电影化信息。此外，我们开发了一个统一的智能体评估系统（UniV-Eval），它标准化了所有任务中的提示、指令解析和评分，从而能够对统一视频模型进行公平、可扩展且可复现的比较。通过将评估建立在基于指令的多镜头视频任务上，UniVBench 提供了首个衡量视频基础模型旨在实现的集成能力的框架。大量人工标注确保我们的评估与人类判断一致，从而能够进行严格的评估并加速朝着稳健视频智能的进展。
英文原文摘要 1769 字符
Video foundation models aim to integrate video understanding, generation, editing, and instruction following within a single framework, making them a central direction for next-generation multimodal systems. However, existing evaluation benchmarks remain fragmented and limited in scope, as they each target a single task, rely on task-specific metrics, and typically use short or simple video clips. As a result, they do not capture the unified capabilities that these models are designed to deliver. To address this gap, we introduce UniVBench, a benchmark purpose-built for evaluating video foundation models across four core abilities: video understanding, video generation, video editing, and a newly proposed task, video reconstruction, which assesses how faithfully a model can reproduce video content it has encountered. Our benchmark substantially expands the complexity of evaluation by incorporating 200 high-quality, diverse and multi-shot videos, each paired with detailed captions, multi-format editing instructions, and reference images. All videos are entirely human-created and carefully validated, offering richer cinematic information than prior benchmarks. In addition, we develop a unified agentic evaluation system (UniV-Eval) that standardizes prompting, instruction parsing, and scoring across all tasks, enabling fair, scalable, and reproducible comparisons of unified video models. By grounding evaluation in instruction-based multi-shot video tasks, UniVBench provides the first framework for measuring the integrated capabilities that video foundation models aim to achieve. Extensive human annotations ensure our evaluation aligns with human judgment, enabling rigorous assessment and accelerating progress toward robust video intelligence.