Evaluating Model Improvement Strategies
A tech startup wants to improve its general-purpose language model for the specific task of summarizing customer support tickets. Two teams propose different methods. Based on the descriptions provided in the case study, which team's proposal involves modifying the model's underlying parameters, and which one enhances performance at the moment of use without such modifications? Justify your answer by explaining the core difference in their approaches.
0
1
Tags
Ch.5 Inference - Foundations of Large Language Models
Foundations of Large Language Models
Foundations of Large Language Models Course
Computing Sciences
Application in Bloom's Taxonomy
Cognitive Psychology
Psychology
Social Science
Empirical Science
Science
Related
Evaluating Model Improvement Strategies
Classifying LLM Scaling Strategies
A development team is using a large, pre-trained language model that is computationally expensive to modify. They need to enhance its performance for a specific, temporary project. A key requirement is that any performance enhancement must be easily removable, restoring the model to its original state without needing to store a separate version. Which scaling approach is most suitable for this scenario?