introduction to ai coursea module 3 answers
Module 3 challenge
1. Which of the following factors can result in hallucinations in large language model (LLM) output? Select three answers.
- High-quality training data
- Varied text-processing techniques
- Lack of contextual understanding
- Unclear prompts
2. A research team uses two different LLMs to generate summaries of published scientific papers. The team notices discrepancies between the summaries that were produced by each LLM. What potential issue with LLM output does this scenario highlight?
- Coding experience is essential to operate LLMs efficiently
- Training differences between LLMs can lead to varied output
- Understanding technical concepts is difficult for most LLMs
- Writing effective summaries is not a strength of LLMs
3. What basic prompt engineering strategies can be used to guide LLMs to generate useful outputs? Select two answers.
- Provide clear and specific instructions
- Repeat details several times
- Include a verb at the beginning of a prompt
- Incorporate friendly banter
4. An advertising executive is leveraging an LLM to suggest slogans for a new drink. Which prompt provides the clearest instructions with the most context?
- Slogans for new drink. Lots of energizing herbs. Really great taste.
- New drink slogans. Mention energizing herbs. Should be better than competitors’ slogans.
- List slogans for a new drink. We want to sell this drink to many customers. Come up with something interesting. Make sure that it’s really catchy.
- Create five potential slogans for a new drink. Use a casual tone. The drink has a tart taste and energizing herbs.
5. A graphic artist writes a prompt asking a large language model to summarize the latest trends in the design industry. Within the prompt, they provide three examples to demonstrate the context, phrasing, and intended format of the output. What does this scenario describe?
- Occasional prompting
- Zero-shot prompting
- Few-shot prompting
- One-shot prompting