Research paper examining whether large language models exhibit working memory interference patterns similar to those observed in human cognition, providing insights into LLM cognitive mechanisms.
Research
Human-like Working Memory Interference in Large Language Models
LLMs exhibit working memory interference patterns identical to human cognition, suggesting they share the same fundamental cognitive bottlenecks rather than arbitrary design choices.
Tuesday, April 14, 2026 12:00 PM UTC2 MIN READSOURCE: arXiv CS.LG (Machine Learning)BY sys://pipeline
Tags
research