Metraipolis › archive › dc_aa6daba7c0ae
Can Vision-Language Models Assess Proxemic Risk from Egocentric Robot Images? — arXiv:2608.12515
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.12515
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T04:55:31.011Z
- SHA-256 of the archived text
8322beea392aa89ddda9c7def66cef6d8d6c3839c429e6f50a5e5b7b0f099300- Stable URL of this page
https://metraipolis.com/documento/dc_aa6daba7c0ae
How to cite
Can Vision-Language Models Assess Proxemic Risk from Egocentric Robot Images? — arXiv:2608.12515 — arXiv (en), https://arxiv.org/abs/2608.12515 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_aa6daba7c0ae (Metraipolis archive dc_aa6daba7c0ae, sha256 8322beea392aa89ddda9c7def66cef6d8d6c3839c429e6f50a5e5b7b0f099300)
BibTeX
@misc{dcaa6daba7c0ae,
title = {Can Vision-Language Models Assess Proxemic Risk from Egocentric Robot Images? — arXiv:2608.12515},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.12515},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_aa6daba7c0ae; sha256 8322beea392aa89ddda9c7def66cef6d8d6c3839c429e6f50a5e5b7b0f099300},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.12515. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Assessing proxemic danger from a robot's egocentric perspective is critical for safe embodied navigation in human environments and requires both visual and contextual reasoning. We evaluate three opensource vision-language models (VLMs) (\textit{InternVL}, \textit{Qwen-VL}, and \textit{SmolVLM}) on the classification of egocentric robot images into four danger levels, comparing three prompting strategies and two rounds of QLoRA fine-tuning against a stratified random baseline. Without fine-tuning, all models perform near the baseline, while fine-tuning yields only modest overall improvements. However, \textit{Qwen-VL} with an advanced prompt achieves substantially higher recall for high-danger cases than the other models. An analysis of person localization further shows that correct danger classification does not correspond to better spatial grounding, indicating that a model may produce a useful safety label without attending to the relevant region of the scene. These results show that current VLMs remain limited in fine-grained proxemic reasoning and spatial grounding, although targeted prompting and fine-tuning can improve high-danger detection in selected models.
Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-14 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2608.12515 — 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 2026-10-19T04:46:30.051Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).