Metraipolis › archive › dc_dc90bf2f585b
Recovering Black-Box Point Cloud Models from Pose Shifts via Derivative-Free Optimization — arXiv:2610.34850
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.34850
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-27T08:02:20.420Z
- SHA-256 of the archived text
41a859a8b10f39c4711d2c9ed9fdf90add98795c1ab5da98c81e63b3b57943ce- Stable URL of this page
https://metraipolis.com/documento/dc_dc90bf2f585b
How to cite
Recovering Black-Box Point Cloud Models from Pose Shifts via Derivative-Free Optimization — arXiv:2610.34850 — arXiv (en), https://arxiv.org/abs/2610.34850 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dc90bf2f585b (Metraipolis archive dc_dc90bf2f585b, sha256 41a859a8b10f39c4711d2c9ed9fdf90add98795c1ab5da98c81e63b3b57943ce)
BibTeX
@misc{dcdc90bf2f585b,
title = {Recovering Black-Box Point Cloud Models from Pose Shifts via Derivative-Free Optimization — arXiv:2610.34850},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.34850},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dc90bf2f585b; sha256 41a859a8b10f39c4711d2c9ed9fdf90add98795c1ab5da98c81e63b3b57943ce},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.34850. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Point cloud models are increasingly deployed through score-only black-box interfaces. Under this access model, test-time rotations outside the training distribution can substantially degrade performance. Although existing mitigations such as equivariant retraining and gradient-based adaptation are effective, they are infeasible without access to model weights or gradients. To address this limitation, we propose CanDO, a test-time canonicalization method that recovers lost performance using only forward queries. Specifically, CanDO searches for an input transformation that maximizes model confidence and uses derivative-free optimization to perform this search without gradients. We further identify three conditions for successful recovery: a training augmentation gap, an accuracy-aligned confidence surrogate, and global search over a multi-modal objective. Across multiple architectures and benchmarks, CanDO recovers 76% of the oracle gap under azimuthal misorientation and 51% under arbitrary SO(3) rotations for object part segmentation. These results show that frozen score-only point cloud models can recover from rotation-induced distribution shifts through post-deployment canonicalization alone. 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); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.34850 — 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 2026-10-27T07:55:30.944Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).