MineAgent: Towards Remote-Sensing Mineral Exploration with Multimodal Large Language Models

TL;DR

MineAgent is a modular framework that enhances multimodal large language models for remote-sensing mineral exploration by improving multi-image reasoning and domain-specific knowledge integration, supported by a new benchmark called MineBench.

Contribution

The paper introduces MineAgent, a novel modular framework for MLLMs in mineral exploration, and MineBench, a benchmark for evaluating domain-specific remote-sensing tasks.

Findings

MineAgent significantly improves multi-image reasoning capabilities.

MineBench effectively evaluates MLLMs in geological and hyperspectral data tasks.

Experiments show MineAgent's potential to advance remote-sensing mineral exploration.

Abstract

Remote-sensing mineral exploration is critical for identifying economically viable mineral deposits, yet it poses significant challenges for multimodal large language models (MLLMs). These include limitations in domain-specific geological knowledge and difficulties in reasoning across multiple remote-sensing images, further exacerbating long-context issues. To address these, we present MineAgent, a modular framework leveraging hierarchical judging and decision-making modules to improve multi-image reasoning and spatial-spectral integration. Complementing this, we propose MineBench, a benchmark specific for evaluating MLLMs in domain-specific mineral exploration tasks using geological and hyperspectral data. Extensive experiments demonstrate the effectiveness of MineAgent, highlighting its potential to advance MLLMs in remote-sensing mineral exploration.