Effective Explanations Support Planning Under Uncertainty
📰 ArXiv cs.AI
Learn how to generate effective explanations that support planning under uncertainty using large language models and planning agents
Action Steps
- Translate explanations into program-like guidance using a large language model
- Execute the guidance using a planning agent under partial observability
- Score explanations by their efficiency in achieving the desired outcome
- Refine explanations based on feedback from the planning agent
- Apply this approach to real-world problems that involve planning under uncertainty
Who Needs to Know This
Researchers and engineers working on AI planning and explanation generation can benefit from this approach to improve the efficiency of their systems
Key Insight
💡 Effective explanations can be generated by converting utterances into action plans using large language models and planning agents
Share This
🤖 Effective explanations can support planning under uncertainty! 📝 Learn how to generate them using large language models and planning agents #AI #Planning #Explainability
Key Takeaways
Learn how to generate effective explanations that support planning under uncertainty using large language models and planning agents
Full Article
Title: Effective Explanations Support Planning Under Uncertainty
Abstract:
arXiv:2605.08406v1 Announce Type: cross Abstract: Explaining how to get from A to B can be challenging. It requires mentally simulating what the listener will do based on what they are told. To capture this process, we propose a computational model that converts utterances into action plans: a large language model translates an explanation into program-like guidance (a policy prior and value map), and a planning agent executes it under partial observability. We score explanations by the efficien
Abstract:
arXiv:2605.08406v1 Announce Type: cross Abstract: Explaining how to get from A to B can be challenging. It requires mentally simulating what the listener will do based on what they are told. To capture this process, we propose a computational model that converts utterances into action plans: a large language model translates an explanation into program-like guidance (a policy prior and value map), and a planning agent executes it under partial observability. We score explanations by the efficien
DeepCamp AI