Metraipolis › archive › dc_1277fd01ce24
EarthJev: Calibrated, Non-Generative Answers to Natural-Language Questions about Sentinel-2 Land Cover — arXiv:2611.02096
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.02096
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T10:36:23.592Z
- SHA-256 of the archived text
922562fde22680c1fa2c9f7f6496589f9b44b8fbcb2a2a0169004a7507e7085a- Stable URL of this page
https://metraipolis.com/documento/dc_1277fd01ce24
How to cite
EarthJev: Calibrated, Non-Generative Answers to Natural-Language Questions about Sentinel-2 Land Cover — arXiv:2611.02096 — arXiv (en), https://arxiv.org/abs/2611.02096 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1277fd01ce24 (Metraipolis archive dc_1277fd01ce24, sha256 922562fde22680c1fa2c9f7f6496589f9b44b8fbcb2a2a0169004a7507e7085a)
BibTeX
@misc{dc1277fd01ce24,
title = {EarthJev: Calibrated, Non-Generative Answers to Natural-Language Questions about Sentinel-2 Land Cover — arXiv:2611.02096},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.02096},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1277fd01ce24; sha256 922562fde22680c1fa2c9f7f6496589f9b44b8fbcb2a2a0169004a7507e7085a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02096. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Vision-language models for remote sensing answer questions about satellite images by generating text: every answer costs a decoding loop, has to be parsed back to an option, and comes without a usable probability. We present EarthJev, a 0.8B-parameter model that answers typed questions about multispectral Sentinel-2 patches without generating any token. All 12 bands enter the vision encoder of Qwen3.5-0.8B through a widened input layer initialised from the RGB kernels, and, following the open OneJev recipe, the answer is read from the logits of the option letters at a single position and trained with cross-entropy plus Brier score. Under a pre-specified analysis plan, we compare it with the same backbone fine-tuned to generate the answer and with three multimodal language models fine-tuned with LoRA, all trained on the same 70,890 questions from 5% of BigEarthNet v2.0 and timed in one job on the same GPU; VQA models that classify over a fixed answer vocabulary are reported as a reference for the cost of answering only the questions seen in training. EarthJev answers a question in 19.9 ms, 4.3 to 8.7 times faster than the multimodal language models and 6.8 times faster than its own backbone used generatively; when all the questions about an image are asked together, it answers them in 42.8 ms, 3.7 to 10.1 times faster than the batched generative models. On 36,028 test questions it reaches 86.0% accuracy (three seeds), 0.6 and 1.2 points below the two 2B multimodal models and 3.5 points below the remote-sensing EarthDial-4B, with answers that are valid by construction and calibrated probabilities (ECE 0.006). Replacing the nine non-RGB bands with their mean costs 4.5 points, so the model uses the multispectral input, and in an ablation training the whole vision encoder raises accuracy to 88.4% at the same cost. Code and model: this https URL Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2611.02096 — 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 2026-11-03T10:29:48.839Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).