Emerging Human-like Strategies for Semantic Memory Foraging in Large Language Models

Researchers analyzed how Large Language Models access semantic memory using the Semantic Fluency Task, finding that LLMs exhibit similar memory foraging patterns to humans. The study reveals convergent and divergent search strategies in LLMs that mirror human cognitive behavior, potentially enabling better human-AI alignment or productive cognitive disalignment.

• →LLMs demonstrate human-like memory foraging strategies when performing semantic fluency tasks.
• →Both convergent and divergent memory search patterns emerge in distinct layers of LLMs, similar to human cognitive processes.
• →The research applies mechanistic interpretability techniques to understand semantic memory access in AI systems.
• →Findings could lead to better cognitive alignment between humans and AI systems.
• →The study suggests potential for enhancing complementary strengths in human-AI interaction through strategic cognitive disalignment.