This repository is a curated collection of resources on "System 2" reasoning in AI, focusing on deliberative, logical thought processes. It targets AI researchers, engineers, and practitioners interested in cognitive architectures, general problem-solving strategies, and self-improvement in AI agents. The primary benefit is a centralized, comprehensive overview of foundational concepts, recent advancements, and relevant tools in this rapidly evolving field.

How It Works

The repository functions as a living bibliography, meticulously categorizing links to papers, books, blog posts, code repositories, and datasets. It covers a broad spectrum of approaches, including established cognitive architectures (SOAR, ACT-R), modern LLM-based agent frameworks (ReAct, AutoGen), novel prompting techniques (Chain-of-Thought, Tree-of-Thoughts), and emerging research areas like neural-algorithmic reasoning and world models. This structured approach allows users to navigate complex topics and identify key research directions.

Quick Start & Requirements

This repository is a collection of links and does not require installation or execution. Users can browse the content directly.

Highlighted Details

• Extensive lists of cognitive architectures, agent frameworks, and prompting techniques.
• Detailed sections on specific research areas like Neural-Algorithmic Reasoning, World Models, and Neuro-Symbolic Approaches.
• Curated lists of datasets, benchmarks, and software tools relevant to AI reasoning.
• Links to numerous open-source replication efforts and related projects.

Maintenance & Community

The project is community-driven, with contributions welcomed via pull requests. It serves as a community hub for sharing and discussing advancements in AI reasoning.

Licensing & Compatibility

The repository itself contains links to external resources, each with its own licensing. The collection itself is likely under a permissive license, but users must verify the licenses of individual linked projects.

Limitations & Caveats

As a curated list, the content's recency and comprehensiveness depend on community contributions. Some linked projects may be experimental, under active development, or have specific hardware/software requirements not detailed here.