Metraipolis › archive › dc_f784dca7b180
RESUME: Recurrent State Updates from Motion and Residual Signals for Efficient Video Language Modeling — arXiv:2609.39563
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.39563
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T14:46:11.930Z
- SHA-256 of the archived text
f6799832e91ed437802b190dfa231beb30806a566e3f1013aeb0a5687c27dc3b- Stable URL of this page
https://metraipolis.com/documento/dc_f784dca7b180
How to cite
RESUME: Recurrent State Updates from Motion and Residual Signals for Efficient Video Language Modeling — arXiv:2609.39563 — arXiv (en), https://arxiv.org/abs/2609.39563 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f784dca7b180 (Metraipolis archive dc_f784dca7b180, sha256 f6799832e91ed437802b190dfa231beb30806a566e3f1013aeb0a5687c27dc3b)
BibTeX
@misc{dcf784dca7b180,
title = {RESUME: Recurrent State Updates from Motion and Residual Signals for Efficient Video Language Modeling — arXiv:2609.39563},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.39563},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f784dca7b180; sha256 f6799832e91ed437802b190dfa231beb30806a566e3f1013aeb0a5687c27dc3b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39563. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Existing video language models encode sampled RGB frames independently, so a long video must either exhaust the token budget or drop the changes between sampled frames. Codec-aware front-ends read the motion vectors and residuals that encoding produced, but in their deployed form each predictive frame is still tokenized on its own: the tokens are a function of the current primitives, not of a carried reference. We argue that a more natural function is of both---the current primitives and a carried reference. A clip and its time reversal share the same frames and differ only in the order of changes---an axis that symmetric pooling discards by construction, and that is non-empty in the frozen vision features VideoLMs use---and the codec recurrence already composes those changes in order against a reference state. We introduce RESUME, a stateful codec representation: an anchor I-frame initializes a compact latent state, each subsequent predictive frame is consumed as an update to that state, and a shared readout exposes VideoLM-compatible tokens from the accumulated state. Codec prediction is thereby kept at the representation level and handed to the language model as a trajectory, not as a set of independent token groups. At the same per-predictive-frame token budget as prior codec-aware methods, a predictive frame enters the language model as a readout of what the front-end already knows, not as an encoding of the current primitives alone. Across ten benchmarks, the gains concentrate on temporal reasoning: on all three temporal benchmarks RESUME improves over both the RGB-frame baseline LLaVA-Video-7B (by 2.8, 5.1, and 3.9 points on TempCompass, TOMATO, and MVBench) and the codec-based baseline CoPE-7B, while staying competitive on general and long-form QA. Frozen-transition tests further show anchor dependence, order sensitivity, and useful rollout behavior beyond the training horizon. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-01 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.39563 — 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 2026-10-07T14:19:00.758Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).