Cognitive Architectures and Neurosymbolic Integration

The field of artificial intelligence is witnessing a significant shift towards integrating cognitive architectures and neurosymbolic approaches to enhance reasoning, decision-making, and perception capabilities. Researchers are exploring the role of cognitive imagination, symbolic reasoning, and neurosymbolic frameworks to develop more human-like intelligence in machines. This trend is evident in the development of novel frameworks, such as those that combine vision-language models with symbolic reasoning to detect cognitive attacks in augmented reality or predict head movements in virtual agents. Noteworthy papers in this area include: A Neurosymbolic Framework for Interpretable Cognitive Attack Detection in AR, which presents a novel approach for detecting cognitive attacks in augmented reality. Don't Forget Imagination!, which highlights the importance of cognitive imagination in human thinking and proposes semantic models as a new approach to simulating cognitive imagination.

Sources

Whither symbols in the era of advanced neural networks?

Don't Forget Imagination!

A tutorial note on collecting simulated data for vision-language-action models

How Does a Virtual Agent Decide Where to Look? - Symbolic Cognitive Reasoning for Embodied Head Rotation

A Neurosymbolic Framework for Interpretable Cognitive Attack Detection in Augmented Reality

JRDB-Reasoning: A Difficulty-Graded Benchmark for Visual Reasoning in Robotics

Reasoning in Computer Vision: Taxonomy, Models, Tasks, and Methodologies

Built with on top of