Amortizing intractable inference in large language models

Understand

Autoregressive large language models (LLMs) compress knowledge from their training data through next-token conditional distributions.

• This limits tractable querying of this knowledge to start-to-end autoregressive sampling.
• However, many tasks of interest -- including sequence continuation, infilling, and other forms of constrained generation -- involve sampling from intractable posterior distributions.
• We address this limitation by using amortized Bayesian inference to sample from these intractable posteriors.