Prompt Engineering
Shaping a model's instructions and context to get the output you actually want.
Überblick
Prompt engineering is the practice of structuring instructions, examples and context so a model produces the desired output without any change to its weights. It solves the problem that a capable model can still fail a task purely because the request was ambiguous, and it is far cheaper to iterate on wording than to fine-tune. It exists as a discipline because model behaviour is sensitive to framing, and because techniques such as giving examples or asking for step-by-step reasoning reliably change results.
Technologiegraph
Wie Prompt Engineering mit dem übrigen Stack zusammenhängt.
Nutzt und hängt ab von
- Jailbreak technique Safety & Alignment Jailbreak based on Prompt Engineering High ✓ source
Genutzt von
- Red Teaming technique Safety & Alignment Prompt Engineering used by Red Teaming High ✓ source
Angewendet auf
- Chain of Thought technique Prompting Chain of Thought used for Prompt Engineering High ✓ source
- DSPy framework Frameworks DSPy used for Prompt Engineering High ✓ source
- Few-Shot Learning technique Prompting Few-Shot Learning used for Prompt Engineering High ✓ source
- In-Context Learning concept Prompting In-Context Learning used for Prompt Engineering High ✓ source
- LLM concept Models Prompt Engineering used for LLM High ✓ source
- Self-Consistency technique Prompting Self-Consistency used for Prompt Engineering High ✓ source
- Tree of Thoughts technique Prompting Tree of Thoughts used for Prompt Engineering High ✓ source
- Zero-Shot concept Prompting Zero-Shot used for Prompt Engineering High ✓ source
Classification
- Context Engineering technique Prompting Prompt Engineering part of Context Engineering High ✓ source