Fine-tuning Gemma 3 on 2,700 LLM-generated geo-captions gives competitive image geolocation and a new MR40k rural benchmark.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
GeoLocSFT: Efficient Visual Geolocation via Supervised Fine-Tuning of Multimodal Foundation Models
Fine-tuning Gemma 3 on 2,700 LLM-generated geo-captions gives competitive image geolocation and a new MR40k rural benchmark.