Metraipolis › archive › dc_ff71399fa90d

SyncPro: Synchronous Reasoning and Responding for Proactive Streaming Video Understanding — arXiv:2610.32132

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.32132
License (as archived)
CC0-1.0
Archived
2026-10-26T19:53:49.613Z
SHA-256 of the archived text
d79ef4dfa07b288d9044d38040d78624cd7a79767e942aa0a94750ca35ae4928
Stable URL of this page
https://metraipolis.com/documento/dc_ff71399fa90d

How to cite

SyncPro: Synchronous Reasoning and Responding for Proactive Streaming Video Understanding — arXiv:2610.32132 — arXiv (en), https://arxiv.org/abs/2610.32132 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ff71399fa90d (Metraipolis archive dc_ff71399fa90d, sha256 d79ef4dfa07b288d9044d38040d78624cd7a79767e942aa0a94750ca35ae4928)
BibTeX
@misc{dcff71399fa90d,
  title = {SyncPro: Synchronous Reasoning and Responding for Proactive Streaming Video Understanding — arXiv:2610.32132},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.32132},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ff71399fa90d; sha256 d79ef4dfa07b288d9044d38040d78624cd7a79767e942aa0a94750ca35ae4928},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.32132. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Proactive streaming video understanding requires assistants to track unfolding events and respond when a user's requested condition is met. To make these decisions, assistants must accumulate relevant evidence and reassess response conditions as new observations arrive. Existing work offers complementary capabilities for this process: proactive video assistants learn when to respond, while streaming video reasoning methods generate scheduled intermediate text to organize observations and support subsequent answers. Together, these capabilities provide a foundation for proactive reasoning, but how to coordinate ongoing reasoning with timely responses remains underexplored. In a direct adaptation, fixed reasoning intervals can trigger updates when nothing relevant has changed or postpone useful updates until the next scheduled step. Sharing one generation path also couples two distinct requirements: preserving evidence for future decisions and producing only the response requested by the user. Moreover, responses generated after reasoning must wait for it to finish, even when sufficient evidence is already available. To address these challenges, we introduce SyncPro, a synchronous dual-stream framework with specialized generation paths for evidence tracking and user-facing responses. Its Inner and Response branches share causal history and activate independently, allowing relevant evidence changes to trigger internal updates and responses to proceed concurrently with ongoing reasoning. We construct 189K dual-stream examples and train SyncPro through supervised fine-tuning followed by trajectory-level reinforcement learning. Experiments across three proactive benchmarks show improved response timing and more accurate answers to user requests.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Computation and Language (cs.CL); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.32132 — 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 2026-10-26T19:45:48.390Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).