Model Discovery Agent: LLM-Assisted Bayesian Experiment Design
K. Murphy · 2026
Ce que cela signifie pour vous
Quand la décision finale repose sur une mécanique déterministe plutôt que sur le modèle lui-même, changer de modèle cesse d'être une source de comportements imprévisibles.
Résumé
Couples an LLM used strictly as a proposer of candidate mechanisms with standard Bayesian machinery: sequential Monte Carlo for posteriors and evidence, and value-of-information to choose the next experiment. Across physics, chemistry and neuroscience benchmarks it recovers the correct mechanism far more data-efficiently than an LLM agent working alone. The most transferable result is a robustness one: because the decision sits in the deterministic layer, accuracy stays between 89% and 94% regardless of which model proposes, whereas the pure LLM agent swings from 26% to 81% depending on the model behind it.
