Metraipolis › archive › dc_f01201c109e2

VLX-VR: An Agentic-Aware Video Reasoning Model — arXiv:2609.09985

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.09985
License (as archived)
CC0-1.0
Archived
2026-10-16T07:53:26.359Z
SHA-256 of the archived text
75b438ec61d5d3e71d51933ca075d5fb63f87304a3b1fea87704e3e03e3dd3ae
Stable URL of this page
https://metraipolis.com/documento/dc_f01201c109e2

How to cite

VLX-VR: An Agentic-Aware Video Reasoning Model — arXiv:2609.09985 — arXiv (en), https://arxiv.org/abs/2609.09985 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f01201c109e2 (Metraipolis archive dc_f01201c109e2, sha256 75b438ec61d5d3e71d51933ca075d5fb63f87304a3b1fea87704e3e03e3dd3ae)
BibTeX
@misc{dcf01201c109e2,
  title = {VLX-VR: An Agentic-Aware Video Reasoning Model — arXiv:2609.09985},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.09985},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f01201c109e2; sha256 75b438ec61d5d3e71d51933ca075d5fb63f87304a3b1fea87704e3e03e3dd3ae},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.09985. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Real-world video understanding requires integrating visual, audio, textual, and temporal evidence distributed across a video. Yet many pipelines use a fixed video context and single-pass inference, limiting adaptive evidence acquisition when observations are incomplete, ambiguous, or conflicting. We present VLX-VR, an agentic-aware video reasoning model trained within a video reasoning framework defined by a Think--Memory--Observation loop. At each step, VLX-VR determines the needed evidence, invokes read_memory or write_memory, incorporates the returned Observation, and decides whether to continue or produce the task output. We train VLX-VR with multimodal data, including videos and agent trajectories, using reinforcement learning to learn evidence acquisition, memory use, and termination. On MINERVA, VLX-VR achieves state-of-the-art performance among the models included in our comparison, with 78.79% accuracy. Under the original three duration groups, its accuracies are 76.70%, 78.73%, and 80.92%, with a cross-duration accuracy variance of 2.97~$\mathrm{pp}^2$. On correctly answered samples, 96.20% of VLX-VR's reasoning traces are consistent with the MINERVA reference reasoning traces and the evidence described by them, while approximately 75.80% of all evaluated samples satisfy both answer correctness and this evidence-grounded trace criterion. These results show strong performance and broadly stable behavior across durations, while counting, state changes, causal reasoning, and spatial perception remain challenging.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-10 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.09985 — 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 2026-10-16T07:24:00.917Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-10?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).