Semantic Organization of Human-Generated Word Associations Is More Efficient Than LLM-Generated Word Associations

Posters

Keyword: Linguistics

Abstract: Word associations generated by humans and large language models were analyzed using a network science approach to provide insights into their semantic organization. Human networks had more efficient semantic organization than LLM networks. Additional analyses of micro-level network structure highlighted similarities and differences in semantic organization across different human cultures.

• Cynthia Siew (Presenting Author)