Human-machine cooperation for semantic feature listing

Kushin Mukherjee, Siddharth Suresh, Timothy T. Rogers

Keywords: concepts, large language models, cognitive science, feature norms, matrix completion

TL;DR: We propose a novel method for combining a learned model of human lexical-semantics from limited data with LLM-generated responses to efficiently generate semantic feature norms for a wide variety of concepts

Abstract: Semantic feature norms — lists of features that concepts do and do not possess — have played a central role in characterizing human conceptual knowledge, but require extensive human labor. Large language models (LLMs) offer a novel avenue for the automatic generation of such feature lists, but are prone to significant error. Here, we present a new method for combining a learned model of human lexical-semantics from limited data with LLM-generated data to efficiently generate high-quality feature norms.