analiza aiqbench
This paper proposes a unified framework for world models that integrates key cognitive functions like motivation and metacognition to bridge the gap between machine capabilities and human-like cognition. By introducing "epistemic world models" for scientific discovery, it identifies critical research gaps in intrinsic drive and self-reflection within current agent architectures. The work offers a theoretical taxonomy essential for advancing next-generation AI systems toward more robust reasoning and autonomous problem-solving.
Human Cognition in Machines: A Unified Perspective of World Models
arXiv:2604.16592v2 Announce Type: replace-cross Abstract: This report of world models distinguishes prior works by the cognitive functions they innovate. Many works claim an almost human-like cognitive capability in their world models. To evaluate these claims requires a proper grounding in first principles from human and machine cognition theory. In moving towards human-like world models we present a conceptual unified framework for world models that fully incorporates all the cognitive functions (i.e., memory, perception, language, reasoning, imagining, motivation, and metacognition) and identify gaps in existing research as a guide for future states of the art. In particular, we find that motivation (especially intrinsic motivation) and metacognition remain drastically under-researched, and we propose concrete directions to address these gaps informed by active inference and global workspace theory. We also introduce epistemic world models, a new category encompassing agent frameworks for scientific discovery that operate over structured knowledge. Our taxonomy, applied to video, embodied, and epistemic world models, suggests research directions where prior taxonomies have not.
Abstract:This report of world models distinguishes prior works by the cognitive functions they innovate. Many works claim an almost human-like cognitive capability in their world models. To evaluate these claims requires a proper grounding in first principles from human and machine cognition theory. In moving towards human-like world models we present a conceptual unified framework for world models that fully incorporates all the cognitive functions (i.e., memory, perception, language, reasoning, imagining, motivation, and metacognition) and identify gaps in existing research as a guide for future states of the art. In particular, we find that motivation (especially intrinsic motivation) and metacognition remain drastically under-researched, and we propose concrete directions to address these gaps informed by active inference and global workspace theory. We also introduce epistemic world models, a new category encompassing agent frameworks for scientific discovery that operate over structured knowledge. Our taxonomy, applied to video, embodied, and epistemic world models, suggests research directions where prior taxonomies have not.
Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Emerging Technologies (cs.ET)
arXiv:2604.16592 [cs.RO]
(or arXiv:2604.16592v2 [cs.RO] for this version)
Text integral sub licența deschisă a publicației · 180 cuvinte
Subiecte conexe
The Cognitive Cost of LLM Writing: A Reading-Time Comparison with Human Text
arXiv · 19 oct. 2026
Cognitive Trajectory Modeling: Quantifying Human-AI Co-Creation through Cognitively Grounded Interaction Trajectories
arXiv · 16 iun. 2026
"very likely" Means "uncertain"? How LLMs Diverge from Humans in Linguistic Uncertainty Quantification
arXiv · 2 oct. 2026
Content Depth Matters in Short-Video Recommendation: Rethinking the Attention Economy
arXiv · 17 aug. 2026