Metraipolis › archive › dc_7bf1256fb471
Pro-Bench: Prompt-Robust Open-Vocabulary Visual Grounding Across Real-World Heterogeneous Environments — arXiv:2609.27076
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.27076
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T12:34:38.250Z
- SHA-256 of the archived text
be6a86654ff24cb94a7cfc7f27e49df8dca5b686643c6e7dbe2724aa729788a3- Stable URL of this page
https://metraipolis.com/documento/dc_7bf1256fb471
How to cite
Pro-Bench: Prompt-Robust Open-Vocabulary Visual Grounding Across Real-World Heterogeneous Environments — arXiv:2609.27076 — arXiv (en), https://arxiv.org/abs/2609.27076 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7bf1256fb471 (Metraipolis archive dc_7bf1256fb471, sha256 be6a86654ff24cb94a7cfc7f27e49df8dca5b686643c6e7dbe2724aa729788a3)
BibTeX
@misc{dc7bf1256fb471,
title = {Pro-Bench: Prompt-Robust Open-Vocabulary Visual Grounding Across Real-World Heterogeneous Environments — arXiv:2609.27076},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.27076},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7bf1256fb471; sha256 be6a86654ff24cb94a7cfc7f27e49df8dca5b686643c6e7dbe2724aa729788a3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.27076. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Open-vocabulary visual grounding enables robots to localise task-relevant entities from natural-language queries without dependence on predefined perceptual taxonomies. However, existing benchmarks largely rely on short category labels and web-scraped imagery, leaving it unclear whether open-vocabulary models can robustly ground diverse queries and visual conditions under real deployments. We introduce \textbf{Pro-Bench}, a prompt-conditioned benchmark for open-vocabulary visual grounding in heterogeneous, real-world environments. Pro-Bench includes $13k+$ RGB frames from independent robotic domains (subterranean, industrial, indoor, outdoor, urban), with $74.5k$ manual instance annotations and $515$ target queries covering categorical, attributive, relational, affordance, state, part-whole, negative, and compositional semantics. We benchmarked $16$ open-vocabulary model configurations in strict zero-shot inference, measuring localisation accuracy across IoU thresholds, end-to-end inference latency, prompt-induced performance variation, and target recovery consistency. Our results show that prompt-robustness is strongly architecture-dependent. Most model configurations ($10/16$) perform best with short category labels, whereas free-form queries yield the highest accuracy for only one. Moreover, similar aggregate mAP can conceal substantial differences in consistent target recovery across reformulations. Pro-Bench enables systematic evaluation of these gaps and supports prompt-robust visual grounding. Pro-Bench: this https URL .
Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-24 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.27076 — 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 2026-10-08T12:12:35.639Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).