aiqbench analysis
This research demonstrates a novel architecture where Large Language Models perform sea surface temperature forecasting by fusing textual environmental data with continuous graph-derived spatial prefixes, avoiding the inefficiency of serializing full grids. By integrating static geographic relations and dynamic SST correlations via Graph Neural Networks, the method achieves superior multi-step prediction accuracy while providing rule-based contextual explanations for trends.
Textual Environmental Context and Spatial Graphs for LLM-Based Regional SST Forecasting
arXiv:2610.07895v1 Announce Type: new Abstract: Sea surface temperature (SST) forecasting depends on local temporal persistence, regional spatial dependence, and environmental conditions that evolve with the forecast date. We study how these heterogeneous conditions can be presented to a large language model (LLM) for regional multi-step forecasting without serializing the full SST grid as text. We formulate forecasting as conditional numerical generation: historical SST and anomaly sequences, date-aligned environmental records, and static ocean knowledge form a textual context, while regional spatial state is supplied through continuous graph-derived prefixes. A static graph encodes persistent geographic--climatological relations, and a dynamic graph encodes recent SST correlations and localized tropical-cyclone influence. Two graph neural networks produce a target-node representation that is mapped by a spatial-prefix fusion and injected into the LLM input. On SST forecasting in the South China Sea, the complete configuration achieves the best MAE and $\Rtwo$ among the compared methods over ten forecast steps. Alongside the numerical forecast, a rule-based module matches predicted trends and environmental-factor directions with knowledge entries to return source-linked, post-hoc contextual explanations.
Abstract:Sea surface temperature (SST) forecasting depends on local temporal persistence, regional spatial dependence, and environmental conditions that evolve with the forecast date. We study how these heterogeneous conditions can be presented to a large language model (LLM) for regional multi-step forecasting without serializing the full SST grid as text. We formulate forecasting as conditional numerical generation: historical SST and anomaly sequences, date-aligned environmental records, and static ocean knowledge form a textual context, while regional spatial state is supplied through continuous graph-derived prefixes. A static graph encodes persistent geographic--climatological relations, and a dynamic graph encodes recent SST correlations and localized tropical-cyclone influence. Two graph neural networks produce a target-node representation that is mapped by a spatial-prefix fusion and injected into the LLM input. On SST forecasting in the South China Sea, the complete configuration achieves the best MAE and $\Rtwo$ among the compared methods over ten forecast steps. Alongside the numerical forecast, a rule-based module matches predicted trends and environmental-factor directions with knowledge entries to return source-linked, post-hoc contextual explanations.
Artificial Intelligence (cs.AI)
arXiv:2610.07895 [cs.AI]
(or arXiv:2610.07895v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2610.07895 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 263 words
Related coverage
PatchSTG: Scalable Spatiotemporal Graph Transformers for Traffic Forecasting on Irregular Sensor Networks
arXiv · Jun 10, 2026
FrED: External Data Influence Estimation via Domain Knowledge Graph Grounding
arXiv · Jul 28, 2026
From PDFs to insights: Architecting an intelligent document processing pipeline with AWS generative AI services
AWS · Jun 12, 2026
Predicting Groundwater Arsenic Concentrations Using Graph Neural Networks
arXiv · Jul 23, 2026