Verifiably grounded machine interpretation of lunar geology

2026-08-10

AI Summary

This study addresses the challenge of achieving verifiable, evidence-based automated reconstruction and dating of lunar mare basaltic stratigraphy. To this end, we propose a novel approach that integrates multimodal vision–language models with coregistered remote sensing data—including topography, spectroscopy, and geological maps—and introduces, for the first time, an open-book retrieval-augmented generation (RAG) mechanism. This framework explicitly links local observational evidence with external scientific literature, clearly distinguishing between qualitative interpretations and quantitative chronologic sources. The system not only accurately characterizes the relationships between lunar stratigraphy and topography but also enables credible age estimation by retrieving published geochronological data, thereby avoiding reliance on model priors and significantly enhancing the data-driven nature and scientific verifiability of geological inferences.

Abstract

Planetary geology relies on historical, interpretive reasoning to reconstruct past events from diverse observations. Here, we present a step toward an automated "machine intelligence geologist" by embedding this distinct methodology of geologic knowledge discovery and inference into a multimodal vision-language architecture. Focusing on the stratigraphy of lunar basaltic mare volcanism, we train a model to generate verifiably grounded geologic interpretations directly from co-registered topographic, spectral, and geologic maps. We demonstrate that while the system successfully balances established geological priors with local visual evidence to accurately describe stratigraphy and terrain, numeric age dating derived solely from vision defaults to memorized priors. Integrating an open-book retrieval mechanism resolves this, enabling the model to faithfully cite published chronologies. Our findings delineate the necessary architecture for automated geologic inference: site evidence must be visually interpreted from local data, while quantitative historical context must be retrieved from the scientific record.

Problem

• lunar geology
• stratigraphy
• machine interpretation
• verifiable grounding
• geologic inference

Innovation

• verifiably grounded interpretation
• multimodal vision-language model
• lunar stratigraphy
• open-book retrieval
• automated geologic inference