Serial Position Effects of Large Language Models

arXiv:2406.15981v1 Announce Type: new
Abstract: Large Language Models (LLMs) have shown remarkable capabilities in zero-shot learning applications, generating responses to queries using only pre-training information without the need for additional fine-tuning. This represents a significant departure from traditional machine learning approaches. Previous research has indicated that LLMs may exhibit serial position effects, such as primacy and recency biases, which are well-documented cognitive biases in human psychology. Our extensive testing across various tasks and models confirms the widespread occurrence of these …

abstract applications arxiv capabilities cs.cl effects fine-tuning information language language models large language large language models llms machine machine learning pre-training queries research responses traditional machine learning training tuning type zero-shot