Summary of Large Language Models As Planning Domain Generators, by James Oswald et al.
Large Language Models as Planning Domain Generators
by James Oswald, Kavitha Srinivas, Harsha Kokel, Junkyu Lee, Michael Katz, Shirin Sohrabi
First submitted to arxiv on: 2 Apr 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper investigates the possibility of using large language models (LLMs) to generate planning domain models from simple textual descriptions. The authors propose a framework for evaluating the quality of LLM-generated domains by comparing plans for different domain instances. An empirical analysis of 7 LLMs across 9 planning domains and 3 types of natural language descriptions is conducted, showing that high-parameter-count LLMs can generate correct planning domains with moderate proficiency. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper explores how to use big language models to make it easier to create plans for different situations. The goal is to see if these models can be used to automatically create plan-making domain models from simple text descriptions. To test this, the authors developed a way to compare the plans made for different scenarios generated by these models. They tested 7 large language models on 9 different planning tasks and found that some of them were able to correctly generate plan-making domains most of the time. |
