Bootstrapping Cognitive Agents with a Large Language Model
arXiv:2403.00810 · doi:10.1609/aaai.v38i1.27822
Abstract
Large language models contain noisy general knowledge of the world, yet are hard to train or fine-tune. On the other hand cognitive architectures have excellent interpretability and are flexible to update but require a lot of manual work to instantiate. In this work, we combine the best of both worlds: bootstrapping a cognitive-based model with the noisy knowledge encoded in large language models. Through an embodied agent doing kitchen tasks, we show that our proposed framework yields better efficiency compared to an agent based entirely on large language models. Our experiments indicate that large language models are a good source of information for cognitive architectures, and the cognitive architecture in turn can verify and update the knowledge of large language models to a specific domain.
References in corpus (26)
- Do As I Can, Not As I Say: Grounding Language in Robotic Affordances
- On Evaluation of Embodied Navigation Agents
- PaLM-E: An Embodied Multimodal Language Model
- TidyBot: Personalized Robot Assistance with Large Language Models
- Voyager: An Open-Ended Embodied Agent with Large Language Models
- Universal and Transferable Adversarial Attacks on Aligned Language Models
- Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents
- Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback
- ChatGPT for Robotics: Design Principles and Model Abilities
- LLM+P: Empowering Large Language Models with Optimal Planning Proficiency
- Large Language Models as Zero-Shot Human Models for Human-Robot Interaction
- Translating Natural Language to Planning Goals with Large-Language Models
- ProgPrompt: Generating Situated Robot Task Plans using Large Language Models
- Code as Policies: Language Model Programs for Embodied Control
- A Survey of Explainable Reinforcement Learning
- Ghost in the Minecraft: Generally Capable Agents for Open-World Environments via Large Language Models with Text-based Knowledge and Memory
- Language Models Meet World Models: Embodied Experiences Enhance Language Models
- Towards A Unified Agent with Foundation Models
- AmadeusGPT: a natural language interface for interactive animal behavioral analysis
- GPT4Vis: What Can GPT-4 Do for Zero-shot Visual Recognition?
- Integrating Diverse Knowledge Sources for Online One-shot Learning of Novel Tasks
- Is Self-Repair a Silver Bullet for Code Generation?
- TIDEE: Tidying Up Novel Rooms using Visuo-Semantic Commonsense Priors
- Improving Knowledge Extraction from LLMs for Task Learning through Agent Analysis
- Towards Modeling and Influencing the Dynamics of Human Learning
- Embodied Executable Policy Learning with Language-based Scene Summarization