Databricks Generative AI Engineer Associate Question 167
Single answerYou are tasked with selecting a pre-trained generative language model for summarizing long legal documents. You are browsing a model hub and come across multiple models with the following metadata in their model cards. Which model should you select based on the task requirements?
- A
Model A: Fine-tuned on a dataset of movie reviews, optimized for sentiment analysis.
- B
Model B: Pre-trained on a diverse corpus of legal documents, with a specific fine-tuning for summarization tasks.
- C
Model C: General-purpose language model trained on a mix of web data, blogs, and news articles, with no fine-tuning for specific tasks.
- D
Model D: Fine-tuned for conversational AI tasks, specifically for generating chatbot responses.
Show answer and explanation
Correct answer: B
Explanation
When selecting a pre-trained model from a model hub or marketplace, it is crucial to evaluate the metadata and model card to ensure the model aligns with your specific task requirements. In this case, Model B is the best choice as it has been both pre-trained on legal documents and fine-tuned for summarization tasks, directly addressing the task at hand.
- A. Incorrect.
Model A is designed for sentiment analysis tasks and is not optimized for summarizing legal documents, making it unsuitable for this task.
- B. Correct.
Model B is specifically pre-trained on legal documents and fine-tuned for summarization tasks, making it the most appropriate choice for summarizing long legal documents.
- C. Incorrect.
Model C is a general-purpose language model without fine-tuning for summarization or legal tasks, making it less effective for this specific task.
- D. Incorrect.
Model D is fine-tuned for conversational AI tasks, which are unrelated to summarizing legal documents, making it an inappropriate choice.