Exploring Plan Space through Conversation: An Agentic Framework for LLM-Mediated Explanations in Planning
📰 ArXiv cs.AI
Learn how to use LLMs to generate explanations for planning tasks and improve human-AI collaboration in sequential decision problems
Action Steps
- Build a conversational interface using LLMs to generate explanations for planning tasks
- Configure the LLM to respond to user questions and provide insights into potential solutions
- Apply the agentic framework to facilitate human-AI collaboration in sequential decision problems
- Test the framework with human planners to evaluate its effectiveness in improving understanding and trust
- Compare the results with traditional planning methods to assess the benefits of LLM-mediated explanations
Who Needs to Know This
AI researchers and planners can benefit from this framework to facilitate iterative reasoning and elicitation processes with human planners, increasing trust and understanding of potential solutions
Key Insight
💡 LLMs can be used to generate explanations for planning tasks, improving human-AI collaboration and trust in sequential decision problems
Share This
🤖💡 Improve human-AI collaboration in planning with LLM-mediated explanations! #LLMs #Planning #AI
Key Takeaways
Learn how to use LLMs to generate explanations for planning tasks and improve human-AI collaboration in sequential decision problems
Full Article
Title: Exploring Plan Space through Conversation: An Agentic Framework for LLM-Mediated Explanations in Planning
Abstract:
arXiv:2603.02070v3 Announce Type: replace Abstract: When automating plan generation for a real-world sequential decision problem, the goal is often not to replace the human planner, but to facilitate an iterative reasoning and elicitation process, where the human's role is to guide the AI planner according to their preferences and expertise. In this context, explanations that respond to users' questions are crucial to improve their understanding of potential solutions and increase their trust in
Abstract:
arXiv:2603.02070v3 Announce Type: replace Abstract: When automating plan generation for a real-world sequential decision problem, the goal is often not to replace the human planner, but to facilitate an iterative reasoning and elicitation process, where the human's role is to guide the AI planner according to their preferences and expertise. In this context, explanations that respond to users' questions are crucial to improve their understanding of potential solutions and increase their trust in
DeepCamp AI