Metraipolis › archive › dc_24bc69bbc2f6
Geometry-Aware Test-Time Learning for Quantitative Spatial Reasoning — arXiv:2609.06004
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.06004
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T15:06:45.960Z
- SHA-256 of the archived text
1bc68ba9196cdf4aa3da6a331d178f190bd8f6be5105dfe29a2cbb6cec08981a- Stable URL of this page
https://metraipolis.com/documento/dc_24bc69bbc2f6
How to cite
Geometry-Aware Test-Time Learning for Quantitative Spatial Reasoning — arXiv:2609.06004 — arXiv (en), https://arxiv.org/abs/2609.06004 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_24bc69bbc2f6 (Metraipolis archive dc_24bc69bbc2f6, sha256 1bc68ba9196cdf4aa3da6a331d178f190bd8f6be5105dfe29a2cbb6cec08981a)
BibTeX
@misc{dc24bc69bbc2f6,
title = {Geometry-Aware Test-Time Learning for Quantitative Spatial Reasoning — arXiv:2609.06004},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.06004},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_24bc69bbc2f6; sha256 1bc68ba9196cdf4aa3da6a331d178f190bd8f6be5105dfe29a2cbb6cec08981a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.06004. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Quantitative spatial reasoning in visual-language models (VLMs) aims to infer spatial distances and directional relationships among objects in 3D space from a 2D image and a natural language query. Despite recent progress, VLM spatial reasoning remains brittle under distribution shifts, largely due to the high cost of 3D supervision. As a result, models often produce inconsistent or contradictory predictions when faced with novel object configurations or rephrased spatial queries, revealing a misalignment between learned representations and underlying geometry. To address this, we propose TTL-SR, a geometry-aware Test-Time Learning framework for quantitative Spatial Reasoning that leverages geometric consistency constraints and unlabeled test data to adapt models to target domains. Specifically, TTL-SR augments the input query with geometrically coupled auxiliary queries, filters unreliable predictions via adaptive geometric triggering to construct structured token-level pseudo-labels, and updates model parameters under a geometry-aware multi-objective loss using only test data. Experimental results demonstrate that TTL-SR significantly boosts spatial reasoning performance, yielding 6.47% and 9.41% accuracy gains for Qwen3-VL-4B-Instruct and SpatialRGPT-VILA-1.5-8B on Q-Spatial-ScanNet dataset, respectively. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.06004 — 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 2026-10-16T14:49:57.033Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).