UniVBench：迈向统一评估视频基础模型
UniVBench: Towards Unified Evaluation for Video Foundation Models
AI总结：
UniVBench通过统一评估系统和多任务视频任务，首次提供衡量视频基础模型综合能力的框架。
摘要：
视频基础模型旨在在一个框架内整合视频理解、生成、编辑和指令遵循，使其成为下一代多模态系统的核心方向。然而，现有的评估基准仍然碎片化且范围有限，因为它们每个都针对单一任务，依赖于任务特定的指标，且通常使用短或简单的视频片段。因此，它们无法捕捉这些模型设计以传达的统一能力。为了解决这一差距，我们引入了UniVBench，这是一个专门用于评估视频基础模型的基准，涵盖四个核心能力：视频理解、视频生成、视频编辑以及新提出的任务视频重建，评估模型能多好地重现其已遇到的视频内容。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，大幅扩展了评估的复杂性。所有视频都是人工创建并经过仔细验证，比先前的基准提供了更丰富的电影信息。此外，我们开发了一个统一的代理评估系统（UniV-Eval），它在所有任务中标准化提示、指令解析和评分，使统一视频模型的公平、可扩展和可重复比较成为可能。通过将评估基于基于指令的多镜头视频任务，UniVBench提供了首个测量视频基础模型旨在实现的综合能力的框架。广泛的真人注释确保我们的评估与人类判断一致，从而实现严格的评估并加速向稳健的视频智能迈进。
英文摘要：
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 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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
