Metraipolis › archive › dc_dd4529ff5e9e

StreamScout: Learning When to Look Deeper for Streaming Video Understanding — arXiv:2609.00291

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.00291
License (as archived)
CC0-1.0
Archived
2026-10-17T07:10:27.889Z
SHA-256 of the archived text
d0f432b340d661cfd401a25e44b52b8df8d0dd04b40315eda9f70b4c97551497
Stable URL of this page
https://metraipolis.com/documento/dc_dd4529ff5e9e

How to cite

StreamScout: Learning When to Look Deeper for Streaming Video Understanding — arXiv:2609.00291 — arXiv (en), https://arxiv.org/abs/2609.00291 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dd4529ff5e9e (Metraipolis archive dc_dd4529ff5e9e, sha256 d0f432b340d661cfd401a25e44b52b8df8d0dd04b40315eda9f70b4c97551497)
BibTeX
@misc{dcdd4529ff5e9e,
  title = {StreamScout: Learning When to Look Deeper for Streaming Video Understanding — arXiv:2609.00291},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.00291},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dd4529ff5e9e; sha256 d0f432b340d661cfd401a25e44b52b8df8d0dd04b40315eda9f70b4c97551497},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.00291. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Streaming video understanding requires answering questions that arrive at arbitrary moments over an unbounded video stream. Existing systems primarily focus on what to retain in a bounded memory, yet access that memory using the same fixed-cost procedure for every query, despite substantial variation in the evidence required. We argue that deciding how deeply to access memory for each query is as important as deciding what the memory should store. To this end, we introduce StreamScout, an adaptive inference framework that maintains only a lightweight textual timeline in context as the stream unfolds. At query time, StreamScout progressively augments the timeline with up to three increasingly informative visual views: a glance at recent frames, a uniform look-back over the past stream, and query-salient retrieval. At each stage, the model answers immediately if the available evidence is sufficient; otherwise, it escalates to the next view. To improve this stop-or-escalate policy, we probe the cascade on an auxiliary set and distill the model's empirical competence boundary into supervision for a lightweight LoRA adaptation, yielding StreamScout-S. We further refine the policy through reinforcement learning, allowing the model to explore stopping behaviors beyond imitation of the distilled decisions, yielding StreamScout-R. Across three backbones and three streaming benchmarks, StreamScout and its variants consistently outperform prior streaming methods while substantially reducing inference cost and token consumption; on OVO-Bench, for instance, StreamScout-S improves Qwen3-VL-8B by 14.65 points while using 59% fewer tokens than uniform sampling and answering in 1.04 s on average.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-02 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.00291 — 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 2026-10-17T07:00:58.897Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).