A team is using a large language model for a complex task that involves exploring multiple possible solution paths, like planning a detailed project with many interdependent steps. They use a simple 'step-by-step' prompting method. The model often commits to a suboptimal path early on and fails to correct its course, leading to an inefficient final plan. This scenario highlights a fundamental limitation of basic prompting for complex reasoning. Which of the following statements best analyzes this limitation and the principle for overcoming it?
0
1
Tags
Ch.5 Inference - 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
Advanced Inference-Scaling Methods for Complex Reasoning
Classification of Methods for Scaling LLM Reasoning
Improving a Language Model's Debugging Performance
A team is using a large language model for a complex task that involves exploring multiple possible solution paths, like planning a detailed project with many interdependent steps. They use a simple 'step-by-step' prompting method. The model often commits to a suboptimal path early on and fails to correct its course, leading to an inefficient final plan. This scenario highlights a fundamental limitation of basic prompting for complex reasoning. Which of the following statements best analyzes this limitation and the principle for overcoming it?
Evaluating Prompting Strategies for Complex Reasoning Tasks