Metraipolis › archive › dc_94fcf6446e09

Can Edge-Deployable Vision-Language Models Identify Species? — arXiv:2609.11916

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.11916
License (as archived)
CC0-1.0
Archived
2026-10-16T05:24:33.536Z
SHA-256 of the archived text
f0bbd619b6e8eb1cdc70addefc9968299702b2b18cc8e6eb1c0e10ae4ea44913
Stable URL of this page
https://metraipolis.com/documento/dc_94fcf6446e09

How to cite

Can Edge-Deployable Vision-Language Models Identify Species? — arXiv:2609.11916 — arXiv (en), https://arxiv.org/abs/2609.11916 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_94fcf6446e09 (Metraipolis archive dc_94fcf6446e09, sha256 f0bbd619b6e8eb1cdc70addefc9968299702b2b18cc8e6eb1c0e10ae4ea44913)
BibTeX
@misc{dc94fcf6446e09,
  title = {Can Edge-Deployable Vision-Language Models Identify Species? — arXiv:2609.11916},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.11916},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_94fcf6446e09; sha256 f0bbd619b6e8eb1cdc70addefc9968299702b2b18cc8e6eb1c0e10ae4ea44913},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.11916. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Camera traps often run in the field on edge hardware with limited or no connectivity, making small, locally-deployable vision-language models (VLMs) -- not frontier-scale ones -- the practically relevant class to evaluate for species identification. We test whether models in this deployment-relevant 2--8B range carry genuine taxonomic knowledge, evaluating four such VLMs (Qwen3-VL 2B/4B/8B, Gemma3 4B) against the domain-specific specialist BioCLIP (300M parameters) on a 96-species task, comparing clean iNaturalist photographs against camera-trap imagery from 6 this http URL collections, on two independently-sampled evaluation sets. All models identify species far above chance, but every model -- general-purpose or specialist -- degrades sharply on field imagery (domain gaps of 9.6--26.6 percentage points, consistent across taxonomic levels and both evaluation sets), indicating the degradation reflects general image legibility rather than fine-grained discrimination failure. BioCLIP substantially outperforms every VLM tested (by 33.2--59.2 percentage points across an expanded 200-image sample for every model) despite its far smaller size, suggesting the gap reflects specialized training data rather than model scale; yet BioCLIP's own domain gap (18.0 points) is statistically indistinguishable from the best VLM's (22.3 points), suggesting the clean-to-field degradation itself is a property of the image-quality shift rather than a general-purpose-model weakness. Under open-set prompting, 5.9--9.6% of responses are syntactically valid but taxonomically nonexistent species names; the relative fabrication-rate ranking across models replicates exactly across both evaluation sets, a more robust finding than any single point estimate.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-11 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.11916 — 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 2026-10-16T05:17:52.061Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).