clarion
Enhancing Computational Cognitive Architectures with LLMs: A Case Study
Computational cognitive architectures are broadly scoped models of the human mind that combine different psychological functionalities (as well as often different computational methods for these different functionalities) into one unified framework. They structure them in a psychologically plausible and validated way. However, such models thus far have only limited computational capabilities, mostly limited by the computational tools and techniques that were adopted. More recently, LLMs have proved to be more capable computationally than any other tools. Thus, in order to deal with both real-world complexity and psychological realism at the same time, incorporating LLMs into cognitive architectures naturally becomes an important task. In the present article, a synergistic combination of the Clarion cognitive architecture and LLMs is discussed as a case study. The implicit-explicit dichotomy that is fundamental to Clarion is leveraged for a seamless integration of Clarion and LLMs. As a result, computational power of LLMs is combined with psychological nicety of Clarion.
- Europe > United Kingdom > England > Oxfordshire > Oxford (0.14)
- Europe > United Kingdom > England > Cambridgeshire > Cambridge (0.14)
- North America > United States > New Jersey > Bergen County > Mahwah (0.04)
- (4 more...)
Roles of LLMs in the Overall Mental Architecture
To better understand existing LLMs, we may examine the human mental (cognitive/psychological) architecture, and its components and structures. Based on psychological, philosophical, and cognitive science literatures, it is argued that, within the human mental architecture, existing LLMs correspond well with implicit mental processes (intuition, instinct, and so on). However, beyond such implicit processes, explicit processes (with better symbolic capabilities) are also present within the human mental architecture, judging from psychological, philosophical, and cognitive science literatures. Various theoretical and empirical issues and questions in this regard are explored. Furthermore, it is argued that existing dual-process computational cognitive architectures (models of the human cognitive/psychological architecture) provide usable frameworks for fundamentally enhancing LLMs by introducing dual processes (both implicit and explicit) and, in the meantime, can also be enhanced by LLMs. The results are synergistic combinations (in several different senses simultaneously).
- Europe > United Kingdom > England > Oxfordshire > Oxford (0.28)
- North America > United States > New Jersey > Bergen County > Mahwah (0.04)
- North America > United States > Massachusetts > Middlesex County > Cambridge (0.04)
- (3 more...)
Can A Cognitive Architecture Fundamentally Enhance LLMs? Or Vice Versa?
The paper discusses what is needed to address the limitations of current LLM-centered AI systems. The paper argues that incorporating insights from human cognition and psychology, as embodied by a computational cognitive architecture, can help develop systems that are more capable, more reliable, and more human-like. It emphasizes the importance of the dual-process architecture and the hybrid neuro-symbolic approach in addressing the limitations of current LLMs. In the opposite direction, the paper also highlights the need for an overhaul of computational cognitive architectures to better reflect advances in AI and computing technology.
- Europe > United Kingdom > England > Oxfordshire > Oxford (0.28)
- Europe > United Kingdom > England > Cambridgeshire > Cambridge (0.14)
- North America > United States > New Jersey > Bergen County > Mahwah (0.04)
- (3 more...)
- Health & Medicine (0.69)
- Education (0.46)
- Information Technology > Artificial Intelligence > Natural Language > Large Language Model (1.00)
- Information Technology > Artificial Intelligence > Cognitive Science > Cognitive Architectures (1.00)
- Information Technology > Artificial Intelligence > Machine Learning > Neural Networks > Deep Learning (0.46)
'The Beast Adjoins' Is Seriously Creepy Sci-Fi
The new anthology The Best American Science Fiction & Fantasy 2021 collects 20 of the best short stories of the year. Series editor John Joseph Adams was particularly impressed with Ted Kosmatka's story "The Beast Adjoins," which presents a fresh take on the idea of an AI uprising. "It's so great," Adams says in Episode 492 of the Geek's Guide to the Galaxy podcast. "It pushes all the sense-of-wonder buttons; it's got all this cool character stuff in there. The story riffs on the Von Neumann-Wigner interpretation of quantum mechanics, positing a future in which advanced AIs are unable to function without humans present. Guest editor Veronica Roth, author of Divergent, found the story extremely creepy. "I reached the part where the machines were using people attached to the front of themselves to keep time moving, and I was like, 'This is revolting.
- North America > Dominican Republic (0.05)
- North America > Cuba (0.05)
Sci-Fi Writers Are Grappling with a Post-Trump Reality
At the 2018 Worldcon, fantasy author N.K. Jemisin became the first person to ever win three consecutive Hugo awards for Best Novel. Given that level of success, science fiction editor John Joseph Adams felt she'd be the perfect guest editor for the latest edition of his anthology series The Best American Science Fiction and Fantasy. "Given that she's clearly the face of the genre at the moment, I thought it would be wonderful to have her as guest editor," Adams says in Episode 342 of the Geek's Guide to the Galaxy podcast. "And thankfully she said yes." Caroline M. Yoachim, whose short story "Carnival Nine" appears in the book, says the 20 stories selected by Jemisin reflect the growing diversity of the fantasy and science fiction field. "One of the things I loved about the book was just the sheer variety of it," Yoachim says.