Diagnosing Context-Loss in a Dialogue Model
Based on the principles of training multi-turn dialogue models, explain the most likely flaw in the model's training objective that would cause this specific pattern of context-loss. Describe how the objective function should be structured to ensure the model properly conditions its responses on the entire preceding conversation.
0
1
Tags
Ch.4 Alignment - 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
Comparison of Training Implementations for Multi-Round Dialogue Models
A team is developing a chatbot for multi-turn conversations. They have a dataset of K-round dialogues, each consisting of a sequence of user inputs (x^k) and desired model responses (y^k). To train the model, they must define an objective function that the model's parameters will be optimized to maximize. Which of the following objective functions correctly represents the goal of making the model generate the entire sequence of desired responses accurately within the conversational context?
Diagnosing Context-Loss in a Dialogue Model
Rationale for Cumulative Objective in Dialogue Models
Dataset-Level Objective for Multi-Round Conversational Models