Metraipolis › archive › dc_593e62f720e3
Motion-as-Prompt: Enhancing Motion Reasoning in Multimodal Large Language Models via Motion-Guided Cross-Frame Visual Prompting — arXiv:2608.11655
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.11655
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T05:16:01.196Z
- SHA-256 of the archived text
47a89659fe625504446d6f45428e859b8da50252001e51443ca05e9d5076d0c9- Stable URL of this page
https://metraipolis.com/documento/dc_593e62f720e3
How to cite
Motion-as-Prompt: Enhancing Motion Reasoning in Multimodal Large Language Models via Motion-Guided Cross-Frame Visual Prompting — arXiv:2608.11655 — arXiv (en), https://arxiv.org/abs/2608.11655 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_593e62f720e3 (Metraipolis archive dc_593e62f720e3, sha256 47a89659fe625504446d6f45428e859b8da50252001e51443ca05e9d5076d0c9)
BibTeX
@misc{dc593e62f720e3,
title = {Motion-as-Prompt: Enhancing Motion Reasoning in Multimodal Large Language Models via Motion-Guided Cross-Frame Visual Prompting — arXiv:2608.11655},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.11655},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_593e62f720e3; sha256 47a89659fe625504446d6f45428e859b8da50252001e51443ca05e9d5076d0c9},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.11655. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Motion-centric video reasoning is fundamental to interactive applications such as robotic manipulation and autonomous navigation. However, multimodal large language models (MLLMs) typically process videos through sparse uniform sampling to control visual-token and attention costs. This strategy may discard critical transitions between sampled frames, limiting reasoning about object movement, collisions, and causal interactions. To mitigate this issue, we propose Motion-as-Prompt (MaP), a track-guided cross-frame visual prompting framework. MaP recovers dense point trajectories, selects motion-informative frames, and marks the trajectories accumulated between consecutive sampled frames directly onto the visual inputs, making otherwise hidden displacement, direction changes, and interactions observable to frozen MLLMs. Experiments on CLEVRER and Something-Something-v2 show that MaP consistently improves average motion-reasoning accuracy, yielding gains of 4.2% and 8.9% for GPT-5.5, respectively. Notably, these improvements are obtained without degrading non-motion understanding, highlighting the robustness of MaP. These results demonstrate that MaP provides a simple and effective solution for enhancing motion-centric video reasoning without model training or architectural modification. Project page: this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-13 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.11655 — 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 2026-10-19T05:08:46.709Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-13?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).