Geo-Expert: Towards Expert-Level Geological Reasoning via Parameter-Efficient Fine-Tuning

• URL: http://arxiv.org/abs/2605.24844v1
• Date: Sun, 24 May 2026 03:28:17 GMT
• Title: Geo-Expert: Towards Expert-Level Geological Reasoning via Parameter-Efficient Fine-Tuning
• Abstract summary: Current AI in Earth sciences predominantly targets surface remote sensing and GIS.<n>We introduce Geo-Expert, a family of parameter-efficient geological LLMs fine-tuned on a custom-curated, high-quality instruction dataset.<n>Our evaluation on a novel domain-specific benchmark, Geo-Eval, reveals that a domain-aligned 8B model can outperform open-weight 70B generalists and proprietary GPT-4o on specialized geological reasoning.
• Score: 8.269639082702934
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: While general-purpose Large Language Models (LLMs) applied to Geology often hallucinate when reasoning about subsurface structures and deep-time evolution, current AI in Earth sciences predominantly targets surface remote sensing and GIS. To bridge this gap, we introduce Geo-Expert, a family of parameter-efficient geological LLMs fine-tuned on a custom-curated, high-quality instruction dataset processed using our custom instruction synthesis pipeline. We investigate the impact of model scaling and architecture by fine-tuning three base models: Qwen3-8B, Qwen3-32B, and Gemma-3-27B, with Low-Rank Adaptation (LoRA) method. Our extensive evaluation on a novel domain-specific benchmark, Geo-Eval, reveals that a domain-aligned 8B model can outperform open-weight 70B generalists and proprietary GPT-4o on specialized geological reasoning, while a 32B variant approaches frontier reasoning models. The optimized 8B model further offers a competitive cost-performance ratio for deployment. This work provides a reproducible recipe for democratizing scientific LLMs and establishes a baseline for geological artificial intelligence.

Related papers

• GeoR-Bench: Evaluating Geoscience Visual Reasoning [102.4693880357753]
  It remains unclear how far current AI systems are from achieving genuine geoscience intelligence.<n>We present textbfGeoR-Bench, a underlineBenchmark for evaluating underlineGeoscience visual underlineReasoning through reasoning informed visual editing tasks.
  arXiv  Detail & Related papers  (2026-05-12T05:13:37Z)
• PRUE: A Practical Recipe for Field Boundary Segmentation at Scale [50.194423500109025]
  We conduct the first systematic evaluation of segmentation and geospatial foundation models (GFMs) for global field boundary delineation.<n>A U-Net semantic segmentation model outperforms instance-based and GFM alternatives on a suite of performance and deployment metrics.<n>Our approach provides a practical framework for reliable, scalable, and reproducible field boundary delineation across model design, training, and inference.
  arXiv  Detail & Related papers  (2026-03-28T02:47:46Z)
• GIAT: A Geologically-Informed Attention Transformer for Lithology Identification [6.315199366662337]
  Geologically-Informed Attention Transformer (GIAT) is a novel framework that fuses data-driven geological priors with the Transformer's attention mechanism.<n>GIAT achieves state-of-the-art performance with an accuracy of up to 95.4%, significantly outperforming existing models.
  arXiv  Detail & Related papers  (2026-03-10T04:02:57Z)
• Predicting California Bearing Ratio with Ensemble and Neural Network Models: A Case Study from Turkiye [0.0]
  The California Bearing Ratio (CBR) is a key geotechnical indicator used to assess the load-bearing capacity of subgrade soils.<n>Traditional tests are often time-consuming, costly, and can be impractical, particularly for large-scale or diverse soil profiles.<n>Recent progress in artificial intelligence, especially machine learning (ML), has enabled data-driven approaches for modeling complex soil behavior with greater speed and precision.<n>This study introduces a comprehensive ML framework for CBR prediction using a dataset of 382 soil samples collected from various geoclimatic regions in Trkiye.
  arXiv  Detail & Related papers  (2025-12-09T08:09:55Z)
• GEO-Bench-2: From Performance to Capability, Rethinking Evaluation in Geospatial AI [52.13138825802668]
  GeoFMs are transforming Earth Observation, but evaluation lacks standardized protocols.<n> GEO-Bench-2 addresses this with a comprehensive framework spanning classification, segmentation, regression, object detection, and instance segmentation.<n>Code, data, and leaderboard for GEO-Bench-2 are publicly released under a permissive license.
  arXiv  Detail & Related papers  (2025-11-19T17:45:02Z)
• GeoBS: Information-Theoretic Quantification of Geographic Bias in AI Models [34.611626290720295]
  We establish an information-theoretic framework for geo-bias evaluation, called GeoBS (Geo-Bias Scores)<n>We propose three novel geo-bias scores that explicitly take intricate spatial factors into consideration.
  arXiv  Detail & Related papers  (2025-09-27T20:07:21Z)
• GeoEvolve: Automating Geospatial Model Discovery via Multi-Agent Large Language Models [49.257706111340134]
  We introduce GeoEvolve, a multi-agent LLM framework that couples evolutionary search with geospatial domain knowledge.<n>We evaluate it on two fundamental and classical tasks: spatial (kriging) and spatial uncertainty.<n>It reduces spatial error (RMSE) by 13-21% and enhances uncertainty estimation performance by 17%.
  arXiv  Detail & Related papers  (2025-09-25T21:03:57Z)
• Geological Everything Model 3D: A Promptable Foundation Model for Unified and Zero-Shot Subsurface Understanding [9.766922279347547]
  Geological Everything Model 3D (GEM) is a unified generative architecture that reformulates tasks as prompt-conditioned inference.<n>GEM achieves zero-shot generalization across tasks with heterogeneous prompt types, without retraining for new tasks or data sources.<n>GEM demonstrates broad applicability across surveys and tasks, including Martian radar stratigraphy analysis, structural interpretation in subduction zones, full seismic stratigraphic interpretation, geobody segmentation, and property modeling.
  arXiv  Detail & Related papers  (2025-07-01T04:14:13Z)
• Foundation Models for Geospatial Reasoning: Assessing Capabilities of Large Language Models in Understanding Geometries and Topological Spatial Relations [4.302482081136594]
  We investigate the extent to which a well-known-text (WKT) representation of geometries and their spatial relations are preserved during spatial reasoning.<n>Our workflow employs three distinct approaches to complete the spatial reasoning tasks for comparison.
  arXiv  Detail & Related papers  (2025-05-22T05:21:31Z)
• PEACE: Empowering Geologic Map Holistic Understanding with MLLMs [64.58959634712215]
  Geologic map, as a fundamental diagram in geology science, provides critical insights into the structure and composition of Earth's subsurface and surface.<n>Despite their significance, current Multimodal Large Language Models (MLLMs) often fall short in geologic map understanding.<n>To quantify this gap, we construct GeoMap-Bench, the first-ever benchmark for evaluating MLLMs in geologic map understanding.
  arXiv  Detail & Related papers  (2025-01-10T18:59:42Z)
• Learning Structures in Earth Observation Data with Gaussian Processes [67.27044745471207]
  This paper reviews the main theoretical GP developments in the field. New algorithms that respect the signal and noise characteristics, that provide feature rankings automatically, and that allow applicability of associated uncertainty intervals are discussed.
  arXiv  Detail & Related papers  (2020-12-22T10:46:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.