WorldTensor dataset harmonizes Earth system data for AI models

Researchers have introduced WorldTensor, a novel harmonized global dataset designed to facilitate the training of foundation models for Earth systems. This dataset integrates a wide array of environmental and socioeconomic variables, aligning them onto a standardized 0.25-degree spatial grid and annual temporal framework. WorldTensor aims to overcome the limitations of existing datasets by incorporating human systems alongside climate, land, and ocean data, enabling the development of more comprehensive multimodal Earth system foundation models. AI

IMPACT Enables training of more comprehensive Earth system foundation models by integrating diverse environmental and socioeconomic data.

RANK_REASON Research paper introducing a new dataset for Earth system foundation models. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• Carlos Rodríguez-Pardo
• CatalyzeX
• CF metadata convention
• DagsHub
• Gotit.pub
• Hugging Face
• IArxiv
• netCDF
• ScienceCast
• WorldTensor

• paper
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →