Geo-Strat-RL: Learning Geological Event Reasoning from Verifiable Tasks

• URL: http://arxiv.org/abs/2606.25000v1
• Date: Tue, 23 Jun 2026 16:20:01 GMT
• Title: Geo-Strat-RL: Learning Geological Event Reasoning from Verifiable Tasks
• Abstract summary: We present Geo-Strat-RL, a synthetic environment that generates stratigraphic observations and compact visible-evidence event histories.<n>We show that RLVR improves geological reconstruction in vision-language models (VLMs), increasing geological content scores on held out stratigraphic diagrams.<n>We further evaluate the same held-out geological histories in a synthetic seismic observation domain by converting the generated scenes into acoustic-impedance-derived amplitude sections.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: To evaluate whether vision-language models can reason about geological histories, it is necessary to construct observations for which the underlying process history is known. Furthermore, reasoning over geological histories is not just a question of recognizing visual patterns, but also of understanding temporal and structural relationships that may be only indirectly visible or highly ambiguous. When ground-truth event histories are not uniquely identifiable or are unavailable, it remains an open challenge to teach models capable of visual reasoning to produce valid geological reconstructions that are consistent with both observed evidence and geological principles. We therefore investigate whether defining a verifiable geological reasoning task can improve geological event reconstruction across observation domains through reinforcement learning with verifiable rewards (RLVR). To this end, we present Geo-Strat-RL, a synthetic environment that generates stratigraphic observations and compact visible-evidence event histories. The environment combines a geological generator with an executable verifier that scores chronology, event identity, deposition, and structural relationships. We show that RLVR improves geological reconstruction in vision-language models (VLMs), increasing geological content scores on held out stratigraphic diagrams. We further evaluate the same held-out geological histories in a synthetic seismic observation domain by converting the generated scenes into acoustic-impedance-derived amplitude sections. In this controlled paired-renderer setting, we present evidence that geological reasoning learned from stratigraphic diagram-domain RLVR training transfers to synthetic seismic representations without seismic-specific training examples, supporting the hypothesis that RLVR can teach reusable geological reasoning concepts across related observation formats.

Related papers

• Verifiably grounded machine interpretation of lunar geology [2.4349098308669594]
  We present a step toward an automated "machine intelligence geologist"<n>We train a model to generate verifiably grounded geologic interpretations from co-registered topographic, spectral, and geologic maps.<n>Our findings delineate the necessary architecture for automated geologic inference.
  arXiv  Detail & Related papers  (2026-08-10T08:32:34Z)
• Towards Fast GNN Surrogates for CO2 Migration in Complex Geological Formations [58.079925240809644]
  We propose an end-to-end graph neural surrogate tailored to CO$$ibility plume migration forecasting in geological storage.<n>The method is evaluated on the SPE11A benchmark, a well-known industry test case designed to assess CO$$ storage scenarios.<n>The proposed model produces competitive forecasts of gas saturation and liquid-phase density, with cumulative errors that remain moderate over extended forecasting horizons.
  arXiv  Detail & Related papers  (2026-06-15T18:19:17Z)
• CoarseSoundNet: Building a reliable model for ecological soundscape analysis [73.44688723989053]
  A soundscape is composed of three types of sound: biophony (sounds made by animals), geophony (natural abiotic sounds) and anthropophony (sounds made by humans)
  arXiv  Detail & Related papers  (2026-05-20T13:18:11Z)
• Earth-o1: A Grid-free Observation-native Atmospheric World Model [81.55899748753434]
  We present Earth-o1, an observation-native atmospheric world model that overcomes structural limitations.<n>By integrating diverse sensor inputs into a unified, grid-free dynamical field, the model autonomously advances the atmospheric state in space and time.<n>We show that this fundamentally distinct paradigm enables direct, real-time forecasting and cross-sensor inference without the overhead of explicit numerical solvers.
  arXiv  Detail & Related papers  (2026-05-07T14:27:48Z)
• 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)
• Towards Unified Vision Language Models for Forest Ecological Analysis in Earth Observation [10.566310652119148]
  REO-Instruct is the first unified benchmark designed for both descriptive and regression tasks in Earth Observation datasets.<n>The dataset integrates co-registered Sentinel-2 and ALOS-2 imagery with structured textual annotations generated and validated through a hybrid human AI pipeline.
  arXiv  Detail & Related papers  (2025-11-20T23:28:30Z)
• History Matching under Uncertainty of Geological Scenarios with Implicit Geological Realism Control with Generative Deep Learning and Graph Convolutions [0.10923877073891446]
  The graph-based variational autoencoder represents an architecture that can handle the uncertainty of different geological scenarios.<n>We offer in-depth analysis of the latent space using tools such as PCA, t-SNE, and TDA to illustrate its structure.
  arXiv  Detail & Related papers  (2025-07-14T12:14:17Z)
• 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)
• SPARTAN: A Sparse Transformer Learning Local Causation [63.29645501232935]
  Causal structures play a central role in world models that flexibly adapt to changes in the environment. We present the SPARse TrANsformer World model (SPARTAN), a Transformer-based world model that learns local causal structures between entities in a scene. By applying sparsity regularisation on the attention pattern between object-factored tokens, SPARTAN identifies sparse local causal models that accurately predict future object states.
  arXiv  Detail & Related papers  (2024-11-11T11:42:48Z)
• Learning Discrete Concepts in Latent Hierarchical Models [73.01229236386148]
  Learning concepts from natural high-dimensional data holds potential in building human-aligned and interpretable machine learning models.<n>We formalize concepts as discrete latent causal variables that are related via a hierarchical causal model.<n>We substantiate our theoretical claims with synthetic data experiments.
  arXiv  Detail & Related papers  (2024-06-01T18:01:03Z)
• Spatiotemporal modeling of European paleoclimate using doubly sparse Gaussian processes [61.31361524229248]
  We build on recent scale sparsetemporal GPs to reduce the computational burden. We successfully employ such a doubly sparse GP to construct a probabilistic model of paleoclimate.
  arXiv  Detail & Related papers  (2022-11-15T14:15:04Z)

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.