Large Discovery Models

Large Discovery Models is an AI framework combining an LLM for candidate generation with a Bayesian surrogate for uncertainty scoring, enabling efficient search over design spaces. It was proposed in the paper "Large Discovery Models: learning where

Two-hop subgraph: this entity, every entity it directly relates to, and every entity those neighbors relate to. Drag a node, scroll to zoom, click to inspect — or click any neighbor and re-center the atlas there.