Competence-gated pooling selectively blends LLM forecasts with priors for event forecasting

UIUC-CS · hf · 2026-09-15

UIUC released a paper on competence-gated pooling of language models and priors for event forecasting.

The method introduces a competence gate that selectively integrates LLM forecasts by estimating the domain-level marginal value of language model predictions over external forecast sources—adopting LLM output only where it adds incremental value—thereby improving hybrid forecasting accuracy.

Original post →

More from Research

Research channel →