LLMs Mirror Human Brain
Large Language Models develop a modular architecture mirroring the human brain, with tasks recruiting overlapping neurons across cognitive domains. This discovery sheds light on the fundamental principles of intelligent systems.
Recent research published on arXiv explores the modular organization of Large Language Models (LLMs), comparing it to the human brain's functional specialization. The study analyzes 46 tasks across four cognitive domains: language, formal reasoning, social reasoning, and physical reasoning. Results show that LLMs develop a modular architecture that mirrors the human brain, with tasks drawing on the same network in humans recruiting overlapping neurons in LLMs. This finding suggests that the modular organization of the human brain may not be an evolutionary accident, but rather a fundamental principle of how intelligent systems are built.
Read also
AI Safety Gap Grows
Open-weight AI models are rapidly advancing, but safety concerns persist. A recent report highlights the gap between AI capabilities and safety measures.
AI Knowledge Graph
Researchers have developed a system to construct knowledge graphs from unstructured documents using AI, addressing issues of inconsistency and duplication. The system utilizes a production extraction layer and ontology-guided extraction to validate the knowledge graph.
OpenAI Disbands Risk Team
OpenAI has disbanded its preparedness team, responsible for assessing and mitigating risks associated with AI models. The team's role was to identify potential risks and develop strategies to address them.
