greedy decoding

Definition

The simplest way for a language model to generate text — at each step it just picks the single most likely next word.

A decoding strategy that selects the argmax token at each step from the model's output distribution; the standard choice for factual QA and the regime in which commitment failures are most visible.

Mentioned in 2 episodes

1. 070
   When Models Know the Answer But Say the Wrong Thing Anyway
2. 026
   What RL Actually Does to Language Models, at the Token Level