Turn-level best-of-N sampling improves LLM agent accuracy by 5-10%
Turn-level best-of-N sampling with generator-selector LLMs yields 5-10% accuracy gains and improved consistency on long-horizon tasks by selecting optimal actions at each step.
Stories per day · 30 days
3 stories, newest first · one card per story
Stories tagged #best-of-n
Holdout Best-of-N: Unbiased Evaluation and Its Cost
Holdout Best-of-N unbiasedly estimates selection reward only when selector depth is below the score count.
Rethinking Personalized Generation: Test-Time Alignment via Factorized Ranking Models
Million-parameter ranking models personalize LLM generation more accurately and cheaply than billion-parameter reward models.