Metraipolis › archive › dc_40cb784c694d

VidOmni-Bench: A Benchmark for Fine-Grained Video Understanding via Spatio-Temporal Event Verification across Complexity and Duration — arXiv:2609.21521

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.21521
License (as archived)
CC0-1.0
Archived
2026-10-10T01:10:31.862Z
SHA-256 of the archived text
8828ca3fa9b7e5e8ca6ec61d509663d53770f056ca661235edb92ea0e0e81544
Stable URL of this page
https://metraipolis.com/documento/dc_40cb784c694d

How to cite

VidOmni-Bench: A Benchmark for Fine-Grained Video Understanding via Spatio-Temporal Event Verification across Complexity and Duration — arXiv:2609.21521 — arXiv (en), https://arxiv.org/abs/2609.21521 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_40cb784c694d (Metraipolis archive dc_40cb784c694d, sha256 8828ca3fa9b7e5e8ca6ec61d509663d53770f056ca661235edb92ea0e0e81544)
BibTeX
@misc{dc40cb784c694d,
  title = {VidOmni-Bench: A Benchmark for Fine-Grained Video Understanding via Spatio-Temporal Event Verification across Complexity and Duration — arXiv:2609.21521},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.21521},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_40cb784c694d; sha256 8828ca3fa9b7e5e8ca6ec61d509663d53770f056ca661235edb92ea0e0e81544},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.21521. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

While Video Large Language Models (Video-LLMs) have recently demonstrated strong performance, reliably evaluating their fine-grained video understanding remains challenging. Existing benchmarks often rely on question answering or ground-truth caption matching, where models may succeed through superficial cues and incomplete annotations. To this end, we introduce VidOmni-Bench, a benchmark that requires models to verify whether each event in dense video captions is supported by the video. VidOmni-Bench consists of 500 videos spanning five complexity types and diverse durations from 4 seconds to 90 minutes. After collecting videos along these axes, we use diverse Video-LLMs to generate dense captions and obtain human-verified sentence-level labels, where sentences containing incorrect events serve as hard negatives for evaluation. Our experiments on VidOmni-Bench reveal three key findings: (i) Video-LLMs frequently generate hallucinated descriptions in dense video captioning; (ii) they also struggle as verifiers, failing to reliably detect plausible but incorrect event descriptions; and (iii) model weaknesses vary across video complexity and duration, revealing diverse, model-specific bottlenecks in current Video-LLMs.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-21 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.21521 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-10T00:23:42.543Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).