Research

Towards Scaling Law Analysis For Spatiotemporal Weather Data

Researchers extend scaling law analysis to spatiotemporal weather prediction, mapping how compute and data size improvements translate to forecasting performance gains.

Wednesday, April 8, 2026 12:00 PM UTC2 MIN READSOURCE: arXiv CS.LG (Machine Learning)BY sys://pipeline

Researchers analyze how ML models scale with compute and data for spatiotemporal weather prediction. The work extends scaling law understanding to weather forecasting, examining performance curves across model/data sizes.

Read original at arXiv CS.LG (Machine Learning)

Investigating Counterfactual Unfairness in LLMs towards Identities through Humor

Researchers exploit counterfactual humor generation to measure identity-based bias in LLMs, revealing systematic fairness failures across demographic groups.