liveBook · Manning

Feedback form

As explained in LLMs in Production

A feedback form is a user-interface component that collects structured or free-form responses from users.

A feedback form gives chatbot users a direct way to report their experience. The responses become useful product data rather than disappearing after a conversation. section 8.1.3

Use the collected feedback to:

  • Improve future training or fine-tuning iterations after cleaning and filtering low-quality or troll responses. section 8.1.3
  • Make decisions about changing models, improving prompts, and investigating edge cases, even without training a model. section 8.1.3

How LLMs in Production covers it

How deep it goes
Defines the feedback form’s role and downstream uses, but defers its fields, storage, moderation, and implementation.
Examples
Uses chatbot experience reports, filtered troll responses, model-switching decisions, prompt improvements, and identified edge cases.
What you can do after
Evaluate chatbot quality and apply cleaned feedback to training, model selection, prompting, or edge-case investigation.
How it connects
Connects feedback with fine-tuning, RLHF, model selection, prompt engineering, and edge-case analysis.
Cover of LLMs in Production
liveBookmanning.com
Read LLMs in Production on liveBook Get the book at manning.com

Related concepts

This explainer was generated by Manning's knowledge graph from the text of LLMs in Production. The section links point to the parts of the book it draws on.