Metraipolis › archive › dc_6ca87d223dea
LAYERSCOPE: A Layerwise Characterization of Video and Multimodal Learned Representations — arXiv:2609.28086
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.28086
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T09:13:18.373Z
- SHA-256 of the archived text
744616489c3c8fb185cd442bef629f12caca5a64d6dcf30e13773965c0daf427- Stable URL of this page
https://metraipolis.com/documento/dc_6ca87d223dea
How to cite
LAYERSCOPE: A Layerwise Characterization of Video and Multimodal Learned Representations — arXiv:2609.28086 — arXiv (en), https://arxiv.org/abs/2609.28086 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6ca87d223dea (Metraipolis archive dc_6ca87d223dea, sha256 744616489c3c8fb185cd442bef629f12caca5a64d6dcf30e13773965c0daf427)
BibTeX
@misc{dc6ca87d223dea,
title = {LAYERSCOPE: A Layerwise Characterization of Video and Multimodal Learned Representations — arXiv:2609.28086},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.28086},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6ca87d223dea; sha256 744616489c3c8fb185cd442bef629f12caca5a64d6dcf30e13773965c0daf427},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.28086. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We propose LAYERSCOPE, a label-free, layerwise framework that aims to characterize a model's learned representations in video and multimodal settings. Evaluating downstream performance using representations from final or intermediate layers typically requires large amounts of labeled data, repeated task-specific evaluations, and substantial computation. To address these limitations, LAYERSCOPE uses local, global, distributional, and correspondence-based geometric metrics to compare layerwise representation structure within and across models without requiring task-specific labels. We evaluate seven architecturally diverse models across video and multimodal classification, clustering, and text-to-video retrieval tasks from MVEB/MVEB+. We find that intermediate-layer representations can outperform final-layer and model-default outputs. We also find that no single geometric metric consistently predicts downstream performance, but note that distinct layerwise geometric signatures emerge across model families. LID shows task-dependent relationships with performance, while RankMe provides the strongest measure for classification and clustering, but is not a universal layer selector. We also find that pairing-aware metrics explain retrieval better than distributional distances alone. LAYERSCOPE therefore offers a framework for comparing representations across models and layers, enabling a more systematic evaluation in video and multimodal settings. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-24 in cs.AI). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.28086 — 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 2026-10-08T08:13:26.746Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).