Jul 2026
Elicitation without Backpropagation: Steering Model Behavior by Optimizing the Latent Posterior
Posterior Prefix Tuning optimizes the parameters of a distribution over hard prompts: it draws prior samples once from the BFT via predictive Monte Carlo (PMC), then estimates the gradient by importance sampling against them.
Garrett Baker, Vinay Pathak, Daniel Murfet et al.
· arXiv.org · 0 citations