# LLaVA-LE: Large Language-and-Vision Assistant for Lunar Exploration

Source: [LLaVA-LE: Large Language-and-Vision Assistant for Lunar Exploration](<https://www.sota2.com/research/paper/69c60f180636e5bf4afc8582>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-28T05:00:04.914Z

Authors: Gokce Inal, Pouyan Navard, Alper Yilmaz

Published: 2026-03-25T18:17:53.000Z

## Abstract

Recent advances in multimodal vision-language models (VLMs) have enabled joint reasoning over visual and textual information, yet their application to planetary science remains largely unexplored. A key hindrance is the absence of large-scale datasets that pair real planetary imagery with detailed scientific descriptions. In this work, we introduce LLaVA-LE (Large Language-and-Vision Assistant for Lunar Exploration), a vision-language model specialized for lunar surface and subsurface characterization. To enable this capability, we curate a new large-scale multimodal lunar dataset, LUCID (LUnar Caption Image Dataset) consisting of 96k high-resolution panchromatic images paired with detailed captions describing lunar terrain characteristics, and 81k question-answer (QA) pairs derived from approximately 20k images in the LUCID dataset. Leveraging this dataset, we fine-tune LLaVA using a two-stage training curriculum: (1) concept alignment for domain-specific terrain description, and (2) instruction-tuned visual question answering. We further design evaluation benchmarks spanning multiple levels of reasoning complexity relevant to lunar terrain analysis. Evaluated against GPT and Gemini judges, LLaVA-LE achieves a 3.3x overall performance gain over Base LLaVA and 2.1x over our Stage 1 model, with a reasoning score of 1.070, exceeding the judge's own reference score, highlighting the effectiveness of domain-specific multimodal data and instruction tuning to advance VLMs in planetary exploration. Code is available at https://github.com/OSUPCVLab/LLaVA-LE.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Visual Question Answering on LUCID-VQA (test)](<https://www.sota2.com/research/sota/visual-question-answering-on-lucid-vqa-test.md>): Visual Question Answering; LUCID-VQA (test); GPT Judge Detailed Score: 0.952.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.24696>)

[PDF](<https://arxiv.org/pdf/2603.24696.pdf>)
