Reproduce “From Manifestations to Cognitive Architectures: a Scalable Framework” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Alfredo Ibias, Guillem Ramirez-Miranda, Enric Guinovart, Eduard Alarcon
A scalable framework advances Artificial General Intelligence by reorienting from optimization methods to a modeling approach, constructing cognitive architecture elements bottom-up from fundamental information processing units. This framework introduces the Self-Projecting Persistence Principle and details a hierarchy from basic information units to complex cognitive functions, aiming to mimic the brain's world-modeling capabilities.
Key insights
- The brain functions as a world modeler, not an optimizer, which provides a conceptual basis for achieving Artificial General Intelligence (AGI).
- Intelligence can emerge from simple, active information primitives called 'Footprints' that self-project their content, effectively filling gaps in partial inputs.
- A seamless, hierarchical organization of information from concrete perceptions to abstract reasoning is crucial for building scalable general intelligence.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.