Metraipolis › archive › dc_026d9ffcf760
Rotated, but How Far? Diagnosing and Improving Object-Rotation Reasoning in VLMs — arXiv:2610.05715
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.05715
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:27:07.878Z
- SHA-256 of the archived text
b60e93cef46d652e4ae877646b2e3a77887974ca7296cf9c9d170cb119aa2ea3- Stable URL of this page
https://metraipolis.com/documento/dc_026d9ffcf760
How to cite
Rotated, but How Far? Diagnosing and Improving Object-Rotation Reasoning in VLMs — arXiv:2610.05715 — arXiv, https://arxiv.org/abs/2610.05715 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_026d9ffcf760 (Metraipolis archive dc_026d9ffcf760, sha256 b60e93cef46d652e4ae877646b2e3a77887974ca7296cf9c9d170cb119aa2ea3)
BibTeX
@misc{dc026d9ffcf760,
title = {Rotated, but How Far? Diagnosing and Improving Object-Rotation Reasoning in VLMs — arXiv:2610.05715},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.05715},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_026d9ffcf760; sha256 b60e93cef46d652e4ae877646b2e3a77887974ca7296cf9c9d170cb119aa2ea3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.05715. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Vision-language models (VLMs) can detect that an object has rotated across views, but cannot reliably tell by how much. We introduce OR-Bench, a fine-grained benchmark for object-rotation reasoning with eight tasks covering rotation detection, rotation magnitude estimation, and multi-view rotation reasoning. Across 12 VLMs, the gap is stark: the strongest models approach 100% accuracy on detection, yet even coarse magnitude estimation is near chance. When asked for exact angles, models place 91.8--100% of their predictions on just $0^\circ$, $90^\circ$, and $180^\circ$, a failure we term canonical-angle collapse. This collapse persists even without visual input. Representation probing shows that missing information is only part of the explanation. Although rotation information becomes less recoverable at finer granularity, substantial coarse-grained information remains, and a simple linear probe outperforms the models' generated answers. This suggests that VLMs underuse rotation information they already encode. We therefore propose RotationCue, a lightweight decoder that recovers coarse rotation information from the VLM's own frozen representations and feeds it back to the model as intermediate textual context. Across three VLMs, RotationCue improves every model--task combination on OR-Bench, raising macro-average accuracy by 7.9--12.6 points while preserving general capabilities. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.05715 — 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 2026-10-06T05:24:06.064Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).