Splitting the problem: prompt chaining
Breaking a large job into separate calls is common advice. The splitting itself gains nothing; what gains is the checkpoint you put at the seams.
4 steps
225 XP
A free account is needed
Start the lesson →
Sources
- Wu, T. et al. 2022 · AI Chains: Transparent and Controllable Human-AI Interaction by Chaining LLM Prompts · CHI 2022
- Khot, T. et al. 2023 · Decomposed Prompting: A Modular Approach for Solving Complex Tasks · ICLR 2023
- Madaan, A. et al. 2023 · Self-Refine: Iterative Refinement with Self-Feedback · NeurIPS 2023
- Huang, J. et al. 2024 · Large Language Models Cannot Self-Correct Reasoning Yet · ICLR 2024
ML Academy · an interactive machine learning course that runs in your browser ·
All lessons