EVE: A Domain-Specific LLM Framework for Earth Intelligence
This addresses the need for tailored AI tools in Earth sciences, offering an incremental improvement with domain-specific adaptations and deployment support.
The paper tackles the problem of developing domain-specialized LLMs for Earth Intelligence by introducing EVE, an open-source framework that includes a 24B model optimized for reasoning and QA, which outperforms comparable models on new benchmarks while maintaining general capabilities.
We introduce Earth Virtual Expert (EVE), the first open-source, end-to-end initiative for developing and deploying domain-specialized LLMs for Earth Intelligence. At its core is EVE-Instruct, a domain-adapted 24B model built on Mistral Small 3.2 and optimized for reasoning and question answering. On newly constructed Earth Observation and Earth Sciences benchmarks, it outperforms comparable models while preserving general capabilities. We release curated training corpora and the first systematic domain-specific evaluation benchmarks, covering MCQA, open-ended QA, and factuality. EVE further integrates RAG and a hallucination-detection pipeline into a production system deployed via API and GUI, supporting 350 pilot users so far. All models, datasets, and code are ready to be released under open licenses as contributions to our field at huggingface.co/eve-esa and github.com/eve-esa.