Metraipolis › archive › dc_29adce22cc3e
Joint Latent Reasoning and Observation for Agentic Video Understanding — arXiv:2610.35229
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.35229
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-27T08:02:35.819Z
- SHA-256 of the archived text
03371d4ae8f702395112d03bcd2d5ed22c999dbc6908237441af5122e9035b79- Stable URL of this page
https://metraipolis.com/documento/dc_29adce22cc3e
How to cite
Joint Latent Reasoning and Observation for Agentic Video Understanding — arXiv:2610.35229 — arXiv (en), https://arxiv.org/abs/2610.35229 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_29adce22cc3e (Metraipolis archive dc_29adce22cc3e, sha256 03371d4ae8f702395112d03bcd2d5ed22c999dbc6908237441af5122e9035b79)
BibTeX
@misc{dc29adce22cc3e,
title = {Joint Latent Reasoning and Observation for Agentic Video Understanding — arXiv:2610.35229},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.35229},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_29adce22cc3e; sha256 03371d4ae8f702395112d03bcd2d5ed22c999dbc6908237441af5122e9035b79},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.35229. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Effective video agents must learn not only how to answer, but also where to look next. Transferring this capability to smaller models is challenging because each observation serves two purposes: supporting an answer and guiding the search for further evidence. Textual descriptions may capture what is currently visible while omitting cues needed for subsequent acquisition. We introduce Latent Agentic Reasoning for Video Analysis (LARVA), a framework that distills multi-round video agents through jointly learned latent reasoning and visual observations. Rather than communicating through captions, an observer passes compact continuous representations to a reasoner that selects the next acquisition action or produces the final answer. Crucially, action and answer supervision propagate through this interface, training the observer to communicate information useful for downstream decisions. Observation reconstruction and reasoning-distribution alignment provide auxiliary supervision from textual teacher trajectories. Across five video-understanding benchmarks, LARVA achieves strong performance with 4B and 7B backbones. On VideoHolmes, Video-MME, and LVBench, the 7B model outperforms a textual CoT student trained on the same teacher trajectories by 5.1, 2.5, and 3.7 percentage points, respectively. These results highlight the value of jointly learning how agents represent evidence and act on it when distilling video-understanding capabilities. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.35229 — 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 2026-10-27T07:55:30.975Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).