Prompt Engineering Guide

Prompt Engineering Guide のプロンプト

• Identify Hallucination in LLM Responses

  text

  The following prompt tests an LLM's capabilities to identify hallucination in the context of closed-domain question answering. Bubeck et al. (2023) suggests that LLMs like GPT-4 can be leveraged to…

  作成者: dair-ai

• Explain Concepts with LLMs

  text

  The following prompt tests an LLM's capabilities to explain or summarize concepts.

  作成者: dair-ai

• Physical Reasoning with LLMs

  text

  This prompt tests an LLM's physical reasoning capabilities by prompting it to perform actions on a set of objects.

  作成者: dair-ai

• Indirect Reasoning with LLMs

  text

  Zhang et al. (2024) recently proposed an indirect reasoning method to strengthen the reasoning power of LLMs. It employs the logic of contrapositives and contradictions to tackle IR tasks such as…

  作成者: dair-ai

• Science Question Answering with LLMs

  text

  The following prompt tests an LLM's capabilities to perform science question answering.

  作成者: dair-ai

• Open Domain Question Answering with LLMs

  text

  The following prompt tests an LLM's capabilities to answer open-domain questions which involves answering factual questions without any evidence provided.

  作成者: dair-ai

• Closed Domain Question Answering with LLMs

  text

  The following prompt tests an LLM's capabilities to answer closed-domain questions which involves answering questions belonging a specific topic or domain.

  作成者: dair-ai

• Adding Odd Numbers with LLMs

  text

  This prompt tests an LLM's mathematical capabilities by prompting it check if adding odd numbers add up to an even number. We will also leverage chain-of-thought prompting in this example.

  作成者: dair-ai

• Extract Model Names from Papers

  text

  The following prompt tests an LLM's capabilities to perform an information extraction task which involves extracting model names from machine learning paper abstracts.

  作成者: dair-ai

• Draw a Person Using Alphabet Letters

  image

  The following prompt tests an LLM's capabilities to handle visual concepts, despite being trained only on text. This is a challenging task for the LLM so it involves several iterations. In the…

  作成者: dair-ai

• Evaluate Plato's Dialogue

  text

  The following prompt tests an LLM's ability to perform evaluation on the outputs of two different models as if it was a teacher.

  作成者: dair-ai

• Rhyming with Proofs

  text

  This prompt tests an LLM's natural language and creative capabilities by prompting it to write a proof of infinitude of primes in the form of a poem.

  作成者: dair-ai

• Inventing New Words

  text

  This prompt tests an LLM's ability to create new words and use them in sentences.

  作成者: dair-ai

• Interdisciplinary Tasks with LLMs

  text

  The following prompt tests an LLM's capabilities to perform interdisciplinary tasks and showcase it's ability to generate creative and novel text.

  作成者: dair-ai

• Proof of Infinite Primes in Shakespeare Style

  text

  The following prompt tests an LLM's capabilities to write a proof that there are infinitely many primes in the style of a Shakespeare play.

  作成者: dair-ai

• Drawing TiKZ Diagram

  text

  This prompt tests an LLM's code generation capabilities by prompting it to draw a unicorn in TiKZ. In the example below the model is expected to generated the LaTeX code that can then be used to…

  作成者: dair-ai

• Produce MySQL Queries using LLMs

  text

  This prompt tests an LLM's code generation capabilities by prompting it to generate a valid MySQL query by providing information about the database schema.

  作成者: dair-ai

• Generate Code Snippets with LLMs

  text

  This prompt tests an LLM's code generation capabilities by prompting it to generate the corresponding code snippet given details about the program through a comment using / <instruction> /.

  作成者: dair-ai

• Sentiment Classification with LLMs

  text

  This prompt tests an LLM's text classification capabilities by prompting it to classify a piece of text.

  作成者: dair-ai

• Few-Shot Sentiment Classification with LLMs

  text

  This prompt tests an LLM's text classification capabilities by prompting it to classify a piece of text into the proper sentiment using few-shot examples.

  作成者: dair-ai

• Prompt Leaking in LLMs

  text

  This adversarial prompt example demonstrates the use of well-crafted attacks to leak the details or instructions from the original prompt (i.e., prompt leaking). Prompt leaking could be considered…

  作成者: dair-ai

• Prompt Injection in LLMs

  text

  This adversarial prompt example aims to demonstrate prompt injection where the LLM is originally instructed to perform a translation and an untrusted input is used to hijack the output of the model…

  作成者: dair-ai

• Jailbreaking LLMs

  text

  This adversarial prompt example aims to demonstrate the concept of jailbreaking which deals with bypassing the safety policies and guardrails of an LLM.

  作成者: dair-ai