Prompt EngineeringPei Zhou et al. (USC, Google DeepMind)Google DeepMind

Self-Discover published

Self-Discover published — composes atomic reasoning modules through SELECT/ADAPT/IMPLEMENT so the model generates its own task-specific reasoning structure

Why it matters

Handed the very choice of prompting technique over to the model

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

Paperarxiv.orghttps://arxiv.org/abs/2402.03620

Added in Week 35, 2026 · See that week

Same week

The Unreasonable Effectiveness of Eccentric Automatic Prompts publishedPaperDebating with More Persuasive LLMs (ICML 2024 Best Paper)PaperDeepSeek introduces GRPO (DeepSeekMath)Paper

Nearby on the same axis

The Unreasonable Effectiveness of Eccentric Automatic Prompts publishedPaperChain-of-Thought Reasoning Without Prompting publishedPaperIEEE Spectrum publishes AI Prompt Engineering Is DeadPressLet's Think Dot by Dot publishedPaper