Metraipolis › archive › dc_cb1fe7fdbf4d
VideoLoop: Looped Working Memory Against Semantic Thrashing in Long-Form Video Agents — arXiv:2609.38119
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.38119
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T18:41:47.892Z
- SHA-256 of the archived text
b3b233acd2a25eb8cae2c7010e40696ba9d5aa8b39e802798d19f559594d8684- Stable URL of this page
https://metraipolis.com/documento/dc_cb1fe7fdbf4d
How to cite
VideoLoop: Looped Working Memory Against Semantic Thrashing in Long-Form Video Agents — arXiv:2609.38119 — arXiv (en), https://arxiv.org/abs/2609.38119 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cb1fe7fdbf4d (Metraipolis archive dc_cb1fe7fdbf4d, sha256 b3b233acd2a25eb8cae2c7010e40696ba9d5aa8b39e802798d19f559594d8684)
BibTeX
@misc{dccb1fe7fdbf4d,
title = {VideoLoop: Looped Working Memory Against Semantic Thrashing in Long-Form Video Agents — arXiv:2609.38119},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.38119},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cb1fe7fdbf4d; sha256 b3b233acd2a25eb8cae2c7010e40696ba9d5aa8b39e802798d19f559594d8684},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38119. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Long-form video understanding requires multimodal agents to iteratively gather evidence over many reasoning steps. However, most existing agentic methods suffer from semantic thrashing: as append-only working memory grows, attention to key evidence collapses, and the agent loses access to what it has already found. First, we provide a structural argument showing that append-only memory can incorporate newly observed target evidence, but cannot remove accumulated noise or prevent ordered context growth without a rewrite operator. Second, motivated by this analysis, we propose VideoLoop, a multimodal agent with two coupled loops. The outer loop reasons over the video and the inner loop, after each step, retrieves artifacts from an unbounded filesystem of past observations and intermediate analysis, and rewrites a bounded working memory. Extensive experiments demonstrate the effectiveness of VideoLoop, which improves four popular LVLM backbones in a plug-and-play manner, with an average gain of 4.2% points over baseline on VideoMME (long). Further analysis of working memory suggests that VideoLoop mitigates semantic thrashing: on the hardest quarter of VideoMME (long) questions, a blind judge that reads only the agent's context answers 81.1% correctly, versus 60.9% for the append-only agent. With Gemini 3.1 Pro, VideoLoop reaches 88.3% on VideoMME (long), 88.8% on VideoMMMU, and 80.9% on LongVideoBench (long). Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-30 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.38119 — 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 2026-10-07T18:04:24.943Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).