@inproceedings{xie-etal-2026-gr1,
    title = "{GR}1: Reinforcement-Enhanced {LLM} for Geoscience Reasoning",
    author = "Xie, Yule  and
      Ding, Jiaxin  and
      Deng, Cheng  and
      Gao, Shiqing  and
      Zhang, Junran  and
      Zhang, Sibo  and
      Wang, Zeyuan  and
      Wu, Ke  and
      Ding, Xin  and
      Fu, Luoyi  and
      Jin, Meng  and
      Wang, Xinbing",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.1140/",
    pages = "22730--22743",
    ISBN = "979-8-89176-395-1",
    abstract = "Reinforcement learning (RL) has recently shown remarkable ability to enhance reasoning in large language models (LLMs), yet its potential in scientific domains beyond mathematics remains largely unexplored. Geoscience questions couple broad factual knowledge with multi-step inference and often rely on visual evidence such as maps, cross-sections, and diagrams, making them a challenging but verifiable testbed for RL-based reasoning. To enable this study, we introduce GeoMC-10K, a dataset of 10,000 geoscience multiple-choice questions spanning physical to human geography and high-school to professional levels; over 30{\%} of the questions are image dependent. To support text-only RL on these multimodal questions, we design GeoM2T, a multi-agent framework that converts multimodal questions into descriptive text while preserving answerability and difficulty. Fine-tuning LLaMA-3.1-8B and Qwen-3-8B with Group Relative Policy Optimization (GRPO), incorporating a factual reward mechanism, yields GR1, which achieves absolute accuracy improvements of 5.9{\%} and 13.3{\%}, respectively, and it generalizes to out-of-distribution geoscience benchmarks. Together, GeoMC-10K, GeoM2T, and GR1 establish a scalable benchmark and baseline for RL-enhanced geoscience reasoning."
}