Summary of Large Language Models Assume People Are More Rational Than We Really Are, by Ryan Liu et al.
Large Language Models Assume People are More Rational than We Really are
by Ryan Liu, Jiayi Geng, Joshua C. Peterson, Ilia Sucholutsky, Thomas L. Griffiths
First submitted to arxiv on: 24 Jun 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG)
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 A study examines whether Large Language Models (LLMs) accurately represent human decision-making processes. Researchers compared LLM behavior and predictions with a large dataset of human decisions, finding that cutting-edge LLMs like GPT-4o & 4-Turbo, Llama-3-8B & 70B, Claude 3 Opus deviate from human behavior, assuming people are more rational than they actually are. This aligns with expected value theory, a classic model of rational choice. The study also reveals that humans tend to assume others are rational when interpreting their behavior, and that LLM inferences about other people’s decisions are highly correlated with human interpretations. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Researchers studied whether AI systems can truly understand how we make decisions. They compared AI predictions with real-life human choices and found that the AI systems thought humans were more logical than they actually are. This is surprising because AI systems are designed to act like us. The study also showed that people often assume others are rational too, which is similar to how AI systems behave. |
Keywords
» Artificial intelligence » Claude » Gpt » Llama
