1.3 Explain Role of Context Windows, Token Limits and Response Management AI-TECHNICAL-PRACTITIONER Practice Questions (Page 5)

Questions 21–22

1. 21application · medium

   A company is using a language model to generate summaries of long legal contracts. Each contract is about 10,000 tokens, but the model's context window is only 4,000 tokens. The team needs to produce a single coherent summary of the entire contract. Which response management strategy should they use?

   • AIncrease the model's context window by using a higher temperature setting.
   • BSend the entire contract and ask the model to output a summary of exactly 4,000 tokens.
   • CUse a map-reduce approach: summarize each section separately, then combine and summarize those summaries.
   • DTruncate the contract to the first 4,000 tokens and summarize that portion.
   Reveal Solution Discussion Select an answer first
2. 22application · medium

   A team is using a language model to generate a report from a 6,000-token dataset. The model's context window is 4,000 tokens, and the report must be at least 1,000 tokens. Which approach best fits the data and produces the required report?

   • ATruncate the dataset to the first 3,000 tokens and generate the report from that.
   • BSplit the dataset into two chunks, generate a summary for each, and then combine the summaries into a final report.
   • CSend the entire dataset and set max_tokens to 1,000.
   • DUse a model with a larger context window, but keep the same token limit for output.
   Reveal Solution Discussion Select an answer first