Metraipolis › archive › dc_fd507898aa09

LiveProBench: Can Streaming Video Models Really Interact Like Humans? — arXiv:2609.12658

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.12658
License (as archived)
CC0-1.0
Archived
2026-10-16T03:15:49.645Z
SHA-256 of the archived text
5e000db6e044226cc4a217018259611d7a866fdd96e2b04fa94c51e9b661e86c
Stable URL of this page
https://metraipolis.com/documento/dc_fd507898aa09

How to cite

LiveProBench: Can Streaming Video Models Really Interact Like Humans? — arXiv:2609.12658 — arXiv (en), https://arxiv.org/abs/2609.12658 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fd507898aa09 (Metraipolis archive dc_fd507898aa09, sha256 5e000db6e044226cc4a217018259611d7a866fdd96e2b04fa94c51e9b661e86c)
BibTeX
@misc{dcfd507898aa09,
  title = {LiveProBench: Can Streaming Video Models Really Interact Like Humans? — arXiv:2609.12658},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.12658},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fd507898aa09; sha256 5e000db6e044226cc4a217018259611d7a866fdd96e2b04fa94c51e9b661e86c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.12658. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Streaming video understanding requires models to process continuous multimodal input while maintaining temporal context. Existing evaluations are predominantly reactive: they query a model at a selected timestamp and therefore do not assess when it should respond. Proactive interaction instead requires monitoring a standing request, responding within an appropriate interval after the target event, and otherwise remaining silent. We introduce LiveProBench, which evaluates models at one-second stream intervals without an explicit response cue. Its six subtasks vary trigger ambiguity and timing tolerance. Event Sensitivity geometrically combines response and silence rates on the same recording; four window-based subtasks distinguish early, in-window, and missed responses; and Duplicate Counting penalizes omissions and repetitions. Premature responses outnumber missed responses for half of the evaluated models, revealing a substantial gap in the temporal decision-making required for human-like interaction.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-14 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.12658 — 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 2026-10-16T03:05:44.055Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).