Architectural Implications for Prompt Comprehension
A developer is building an application that requires generating long-form text based on a detailed, multi-sentence prompt. They need the model to have the most comprehensive understanding of the entire prompt before it begins generating the response. They are considering two types of models:
- A model that processes the prompt tokens one by one, where each token can only attend to the ones that came before it.
- A model that can process the entire prompt at once, allowing every token in the prompt to attend to every other token in the prompt, before starting to generate the new text.
Which of these two approaches is better suited for the developer's goal? Justify your answer by explaining the fundamental difference in how each model builds its initial understanding of the prompt.
0
1
Tags
Ch.1 Pre-training - Foundations of Large Language Models
Foundations of Large Language Models
Foundations of Large Language Models Course
Computing Sciences
Analysis in Bloom's Taxonomy
Cognitive Psychology
Psychology
Social Science
Empirical Science
Science
Related
A language model is given the prompt 'Despite the initial positive reviews, the film's box office performance was ultimately disappointing because...' and is tasked with generating a continuation. Consider two different ways the model could process this prompt before generating the next token:
- Method 1: When processing the prompt, the token 'disappointing' can directly see and incorporate information from the token 'positive' at the beginning of the sentence.
- Method 2: When processing the prompt, the token 'disappointing' can only see and incorporate information from the preceding tokens, such as 'ultimately' and 'was'.
Which of the following statements best analyzes the fundamental difference in how these two methods build an understanding of the prompt?
Architectural Implications for Prompt Comprehension
Architectural Choice for a Conversational AI