Learn Before
Refining a Specialized Legal LLM
A team has fine-tuned a large language model to summarize legal clauses from contracts. Despite this training, the model frequently includes extraneous information, such as historical context or related legal precedents, instead of sticking strictly to the summarization task. The team proposes two strategies for an additional round of training to correct this behavior. Evaluate the two strategies and justify which one is more likely to be effective.
0
1
Tags
Ch.4 Alignment - Foundations of Large Language Models
Foundations of Large Language Models
Foundations of Large Language Models Course
Computing Sciences
Evaluation in Bloom's Taxonomy
Cognitive Psychology
Psychology
Social Science
Empirical Science
Science
Related
A development team has trained a language model to function as a specialized chatbot for booking restaurant reservations. After the initial training, they find that the model often answers questions about recipes or restaurant reviews, deviating from its core task. Which of the following strategies is most likely to effectively steer the model back to its intended specialized function?
Refining a Specialized Legal LLM
Refining a Specialized Code Generation Model