Are Video Models Ready as Zero-Shot Reasoners? An Empirical Study with the MME-CoF Benchmark
Dataset Analysis
This paper introduces MME-CoF, a benchmark evaluating Veo-3's zero-shot visual reasoning across 12 dimensions, finding video models are not yet reliable standalone reasoners.
Provenance
Collected from papers.
Derived from paper: Are Video Models Ready as Zero-Shot Reasoners? An Empirical Study with the MME-CoF Benchmark