Prompt EngineeringChengrun Yang et al. (Google DeepMind)Google DeepMind

OPRO published

Made by AIWritten by AI (Claude) from the source below. Check anything important against the source.How it's made

OPRO published — uses the LLM as an optimizer, accumulating previous solutions and their scores in the prompt to generate new instructions

Why it matters

Formalized automatic prompt optimization as a general optimization problem

Source — Primary sources first; secondary reports are labelled “Press”

Paperarxiv.orghttps://arxiv.org/abs/2309.03409

Added in Week 35, 2026 · See that week

On the map — ① Prompt Engineering line

  1. Previous stopPaLM 2
  2. This recordOPRO
  3. Next stopChain-of-Verification
See it on the map

Same week

Modular makes Mojo available for local downloadOfficial

Nearby on the same axis

Chain-of-Verification publishedPaperGraph of Thoughts publishedPaperPromptbreeder publishedPaperAnalogical Prompting publishedPaper