Memory GAPS: Would LLMs pass the Tulving Test?

TL;DR

This paper explores whether the Tulving Test, a classic memory assessment, can be used to evaluate the memory capabilities of large language models, linking human memory theories to AI performance.

Contribution

It introduces a novel approach to assess LLMs' memory using the Tulving Test, bridging human memory models and artificial intelligence evaluation.

Findings

LLMs' recognition and recall performance analyzed

Insights into LLMs' memory capabilities relative to human models

Potential limitations of LLMs in memory tasks identified

Abstract

The Tulving Test was designed to investigate memory performance in recognition and recall tasks. Its results help assess the relevance of the "Synergistic Ecphory Model" of memory and similar RK paradigms in human performance. This paper starts investigating whether the more than forty-year-old framework sheds some light on LLMs' acts of remembering.