Summary of Codenames As a Benchmark For Large Language Models, by Matthew Stephenson et al.
Codenames as a Benchmark for Large Language Models
by Matthew Stephenson, Matthew Sidji, Benoît Ronval
First submitted to arxiv on: 16 Dec 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Computation and Language (cs.CL)
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 proposed benchmark uses Codenames, a popular word-based board game, to evaluate the reasoning capabilities of Large Language Models (LLMs). This challenge requires understanding language, theory of mind, and epistemic reasoning. While LLMs excel in language-based tasks, they struggle with lateral thinking challenges. The paper evaluates state-of-the-art LLMs, including GPT-4o, Gemini 1.5, Claude 3.5 Sonnet, and Llama 3.1, across various board setups. Results show that different models perform better at specific roles and exhibit emergent behaviors. Additionally, the study demonstrates that cooperative LLM agents are more generalizable to a wider range of teammates than prior techniques. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper uses a fun board game called Codenames to test how well computers can think. It’s like trying to figure out which words match clues without knowing what they mean! The computer models, or Large Language Models (LLMs), are really good at understanding language, but struggle with thinking outside the box. Researchers tested many LLMs and found that each one is better at certain things than others. They even tried teams of computers playing together and discovered that some combinations work better than others! |
Keywords
» Artificial intelligence » Claude » Gemini » Gpt » Llama
