A Generative AI Engineer is building an LLM to generate article summaries in the form of a type of poem, such as a haiku, given the article content. However, the initial output from the LLM does not…

Question

Options

• AProvide the LLM with a prompt that explicitly instructs it to generate text in the desired tone and
• BUse a neutralizer to normalize the tone and style of the underlying documents
• CInclude few-shot examples in the prompt to the LLM
• DFine-tune the LLM on a dataset of desired tone and style

• A
  5% (3)
• B
  81% (47)
• C
  12% (7)
• D
  2% (1)

Explanation

A 'neutralizer' that normalizes the tone and style of the source documents (option B) would actively work against the goal. The task is to generate summaries in a specific stylistic form (e.g., haiku), so stripping tone and style from the input makes the LLM's job harder, not easier - it removes potentially useful stylistic signal. The correct approach is to guide the LLM's OUTPUT style, not flatten the INPUT content. Options A (explicit tone/style instruction in prompt), C (few-shot examples), and D (fine-tuning on desired style) all directly address output tone and style and are all valid improvement strategies.

No community discussion yet for this question.