This repository is a curated list of papers, codebases, and datasets focused on decision-making using foundation models, particularly Large Language Models (LLMs) and Vision-Language Models (VLMs). It serves researchers and practitioners in AI, robotics, and reinforcement learning by providing a structured overview of the rapidly evolving field of LLM-driven agents and embodied AI.

How It Works

The repository categorizes resources into key areas: Foundation Models as World Models, Reward Models, Agent Models, and Representation Encoders, alongside specific applications in Multi-modal Decision Making and Embodied AI. It meticulously lists influential papers with direct links to their publications and associated code repositories, offering a comprehensive knowledge base for understanding current research trends and implementations.

Quick Start & Requirements

This is a curated list, not a runnable codebase. No installation or specific requirements are needed to browse the content. Links to papers and codebases are provided for further exploration.

Highlighted Details

• Comprehensive coverage of papers and codebases for LLM-driven decision-making.
• Categorization includes World Models, Reward Models, Agent Models, and Embodied AI.
• Links to key benchmarks like Meta-World, RLBench, and AI2-THOR are provided.
• Includes foundational papers on LLM agents like Voyager, Reflexion, and ReAct.

Maintenance & Community

The repository is maintained by 123penny123 and welcomes contributions via pull requests or direct contact for corrections.

Licensing & Compatibility

The repository itself is a list and does not impose licensing restrictions. Individual linked papers and codebases retain their original licenses.

Limitations & Caveats

As a curated list, it reflects the state of research at the time of its last update and may not include the very latest publications or code. The accuracy and completeness of linked resources are dependent on the original sources.