Metraipolis › archive › dc_90ed42359cf0
Aggregating Visual Information with Optimal Transport for VideoLM Token Compression — arXiv:2608.20473
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.20473
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T10:49:22.458Z
- SHA-256 of the archived text
0b5c99c037ea42c548b04c99ca8d981a15d0a22c6080520e0ecf855a12d1fe91- Stable URL of this page
https://metraipolis.com/documento/dc_90ed42359cf0
How to cite
Aggregating Visual Information with Optimal Transport for VideoLM Token Compression — arXiv:2608.20473 — arXiv (en), https://arxiv.org/abs/2608.20473 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_90ed42359cf0 (Metraipolis archive dc_90ed42359cf0, sha256 0b5c99c037ea42c548b04c99ca8d981a15d0a22c6080520e0ecf855a12d1fe91)
BibTeX
@misc{dc90ed42359cf0,
title = {Aggregating Visual Information with Optimal Transport for VideoLM Token Compression — arXiv:2608.20473},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.20473},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_90ed42359cf0; sha256 0b5c99c037ea42c548b04c99ca8d981a15d0a22c6080520e0ecf855a12d1fe91},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.20473. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Video language models process videos as dense visual-token sequences with substantial representational redundancy. Compressing these sequences is therefore essential for reducing the visual-token burden on language-model decoding. The central challenge is to preserve visual information dispersed across frames under such compression. To this end, we introduce Aggregating Visual Information with Optimal Transport (AVIOT), which casts video token compression as transporting a dense empirical measure of frame observations onto a compact target measure. The resulting source-to-target coupling induces a distribution over source observations for each target support, directly specifying how the compressed video representation is constructed. We further adapt this construction along task and spatial axes. Question conditioning modulates the transport cost between source frames and target supports, while influencing how many supports are allocated to each temporal segment, thereby directing representation capacity toward question-relevant content. At multiple spatial granularities, AVIOT computes region-specific temporal transport plans and adaptively fuses the representations they yield, allowing different regions within the same compact representation to draw from different moments. Evaluations across varying compression ratios show that AVIOT matches or outperforms the uncompressed baseline on multiple video-understanding benchmarks while retaining strong performance at higher compression ratios. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-24 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2608.20473 — 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 2026-10-18T10:48:13.540Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).