Metraipolis › archive › dc_08748e9d671d

MVVBench: Benchmarking 4D Reasoning in Vision-Language Models — arXiv:2609.30952

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.30952
License (as archived)
CC0-1.0
Archived
2026-10-08T01:39:37.206Z
SHA-256 of the archived text
2ad0e175b20c8285a2d7cf4cdc9f5317c6cbcb79f6f828b996b897b497b71e6c
Stable URL of this page
https://metraipolis.com/documento/dc_08748e9d671d

How to cite

MVVBench: Benchmarking 4D Reasoning in Vision-Language Models — arXiv:2609.30952 — arXiv (en), https://arxiv.org/abs/2609.30952 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_08748e9d671d (Metraipolis archive dc_08748e9d671d, sha256 2ad0e175b20c8285a2d7cf4cdc9f5317c6cbcb79f6f828b996b897b497b71e6c)
BibTeX
@misc{dc08748e9d671d,
  title = {MVVBench: Benchmarking 4D Reasoning in Vision-Language Models — arXiv:2609.30952},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.30952},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_08748e9d671d; sha256 2ad0e175b20c8285a2d7cf4cdc9f5317c6cbcb79f6f828b996b897b497b71e6c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.30952. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multi-view video understanding requires integrating spatial and temporal evidence across multiple, often non-overlapping camera streams: tracking entities as they transition between viewpoints, aligning events across time, and reasoning about latent 4D continuity rather than any single visible frame. We introduce MVVBench, a benchmark for multi-view video reasoning built from real world multi camera datasets. Questions are curated to be monocular-ambiguous along both the view and the temporal axis: each question is unanswerable from any single view in the designated input set, and the majority are further unanswerable from any single moment. Each question becomes uniquely solvable only by jointly reasoning across views and across time. MVVBench spans diverse dynamic scenes and probes six capabilities: implicit/explicit attribute identification, implicit/explicit relative distance, relative camera pose, and compositional counting, with human-authored QA and rigorous verification. Beyond benchmarking, we provide an extensive analysis of when and why current vision language models succeed or fail, characterizing errors due to temporal mis-localization, cross-view identity breaks, and brittle multi-hop reasoning. We then study inference-time elicitation strategies that unlock latent multi-view competence---task-specific chain-of-thought scaffolds and structured cross-view evidence aggregation---yielding substantial gains without retraining. Finally, we present preliminary evidence that reinforcement learning with verifiable rewards can elicit some latent multi-view competence in the base model, pointing to training-time approaches as a promising direction for future work. Together, MVVBench offers a rigorous evaluation of 4D multi-view reasoning and a foundation for future progress toward reliable embodied perception.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.30952 — 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 2026-10-08T01:12:14.510Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).