VGI-BENCH: Probing Visual Intelligence in Video Generation Models

2026-08-19

AI Summary

为评估视频生成模型的视觉推理能力，研究引入了VGI-bench基准测试，包含27项任务，用以挑战并分析现有模型在零样本视觉推理中的表现与局限。

Abstract

Recent studies suggest that video generation models can exhibit certain forms of zero-shot visual reasoning through generated frames. Yet reliable evaluation remains challenging: benchmarks should adopt inputs aligned with the visual priors of current video models, require valid evolving processes rather than only plausible final states, and calibrate task difficulty to remain challenging yet partly feasible. To this end, we introduce VGI-bench, containing 27 tasks and 810 instances, organized by a two-level taxonomy of task domains and skill tags for fine-grained evaluation of visual reasoning capabilities of video generation models. Our evaluations show that current generative systems can solve a subset of visually grounded reasoning tasks, but remain far from reliable, with even the strongest model, Seedance~2.0, achieving only 51.0% under our evaluation criteria. Our analysis further explore the output failure modes, input condition sensitivity, performance transfer boundary from synthetic fine-tuning, and internal denoising perspective revealing limited self-correction, where later steps mainly refine early hypotheses rather than correct reasoning errors. We hope VGI-bench will help stimulate the development of next-generation video generation models. We will release our code and data.

Problem

• video generation models
• visual reasoning
• evaluation benchmarks

Innovation

• VGI-bench
• visual reasoning
• video generation models
• benchmarking
• performance evaluation