New MuSS dataset and benchmark tackle multi-shot video generation challenges

By PulseAugur Editorial · Summary by gemini-2.5-flash-lite from 1 source

Researchers have introduced MuSS, a new large-scale dataset designed to improve multi-shot subject-to-video generation. The dataset, derived from over 3,000 movies, addresses challenges in narrative logic, spatiotemporal alignment, and the common AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

Read on arXiv cs.CV →

COVERAGE [1]

arXiv cs.CV TIER_1 · Haojie Zhang, Di Wu, Bingyan Liu, Linjie Zhong, Yuancheng Wei, Xingsong Ye, Nanqing Liu, Yaling Liang · 2026-04-28 04:00

MuSS: A Large-Scale Dataset and Cinematic Narrative Benchmark for Multi-Shot Subject-to-Video Generation

arXiv:2604.23789v1 Announce Type: new Abstract: While video foundation models excel at single-shot generation, real-world cinematic storytelling inherently relies on complex multi-shot sequencing. Further progress is constrained by the absence of datasets that address three core …

COVERAGE [1]

MuSS: A Large-Scale Dataset and Cinematic Narrative Benchmark for Multi-Shot Subject-to-Video Generation