Summary of Puzzle Solving Using Reasoning Of Large Language Models: a Survey, by Panagiotis Giadikiaroglou et al.
Puzzle Solving using Reasoning of Large Language Models: A Survey
by Panagiotis Giadikiaroglou, Maria Lymperaiou, Giorgos Filandrianos, Giorgos Stamou
First submitted to arxiv on: 17 Feb 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 This research paper surveys the abilities of Large Language Models (LLMs) in solving puzzles, providing crucial insights into their potential and limitations in artificial intelligence. The study employs a unique taxonomy dividing puzzles into rule-based and rule-less categories, assessing LLMs’ capabilities through various methods including prompting techniques, neuro-symbolic approaches, and fine-tuning. By reviewing relevant datasets and benchmarks, the researchers evaluate LLMs’ performance, identifying significant challenges in complex puzzle scenarios. The findings highlight the disparity between LLM capabilities and human-like reasoning, particularly in tasks requiring advanced logical inference. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study looks at how good computers are at solving puzzles using Large Language Models (LLMs). It’s a big deal because it can help us understand if these models will be able to solve problems like humans do. The researchers created a special way to group puzzles into two types: those with rules and those without. They tested LLMs in different ways, like giving them hints or letting them figure things out on their own. Then they looked at how well the computers did compared to what we know about human problem-solving. What they found is that computers are pretty good at solving some puzzles, but struggle when it comes to complex problems. |
Keywords
» Artificial intelligence » Fine tuning » Inference » Prompting
