Metraipolis › archive › dc_729199c01cd3

SAFIRE: Safety-Critical Benchmark for Fine-grained Fire and Smoke Understanding in Multimodal LLMs — arXiv:2609.07823

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.07823
License (as archived)
CC0-1.0
Archived
2026-10-16T12:57:50.419Z
SHA-256 of the archived text
0b22a820d0b48770bf37897015fda8d1750e781e6dddeb4f64f37694458c3a68
Stable URL of this page
https://metraipolis.com/documento/dc_729199c01cd3

How to cite

SAFIRE: Safety-Critical Benchmark for Fine-grained Fire and Smoke Understanding in Multimodal LLMs — arXiv:2609.07823 — arXiv (en), https://arxiv.org/abs/2609.07823 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_729199c01cd3 (Metraipolis archive dc_729199c01cd3, sha256 0b22a820d0b48770bf37897015fda8d1750e781e6dddeb4f64f37694458c3a68)
BibTeX
@misc{dc729199c01cd3,
  title = {SAFIRE: Safety-Critical Benchmark for Fine-grained Fire and Smoke Understanding in Multimodal LLMs — arXiv:2609.07823},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.07823},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_729199c01cd3; sha256 0b22a820d0b48770bf37897015fda8d1750e781e6dddeb4f64f37694458c3a68},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.07823. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multimodal Large Language Models (MLLMs) show strong progress on vision-language tasks, yet their reliability in safety-critical settings remains underexplored. Fire-smoke understanding is central to public safety and disaster response, but most existing benchmarks lack diverse real-world scenarios and context-aware evaluation. We introduce SAFIRE, a large-scale benchmark for fire-smoke understanding in MLLMs, comprising 83K captioned images from 20 scenarios and 193K multiple-choice VQA (MCVQA) generated from a 9.7K-image subset, spanning 10 evaluation dimensions from basic perception to higher-order reasoning. A GPT-5.4-assisted multi-stage verification pipeline with MLLM majority voting ensures annotation quality. Evaluating ten open-source MLLMs (8B-38B) yields an average accuracy of 61.9%, exposing major gaps in safety-critical reasoning. We further show that adapting vision encoders with only 7% of our domain-specific data boosts fire-scene classification accuracy from 20.1% to 64.5%, indicating that carefully curated data can yield substantial gains even when data volume is limited. All datasets, models, and code are available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CL).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.07823 — 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 2026-10-16T12:30:25.333Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).