Towards Measuring the Representation of Subjective Global Opinions in Language Models
By Esin Durmus, Karina Nyugen, Thomas Liao, Nicholas Schiefer, Amanda Askell, Anton Bakhtin, Carol Chen, Zac Hatfield-Dodds, Danny Hernandez, Nicholas Joseph and 8 more. arXiv.org, 2023
451 citations recorded by the source corpus.
Large language models (LLMs) may not equitably represent diverse global perspectives on societal issues. In this paper, we develop a quantitative framework to evaluate whose opinions model-generated responses are more similar to. We first build a dataset, GlobalOpinionQA, comprised of questions and answers from cross-national surveys designed to capture diverse opinions on global issues across different countries. Next, we define a metric that quantifies the similarity between LLM-generated survey responses and human responses, conditioned on country. With our framework, we run three experiments on an LLM trained to be helpful, honest, and harmless with Constitutional AI. By default, LLM responses tend to be more similar to the opinions of certain populations, such as those from the USA, and some European and South American countries, highlighting the potential for biases. When we prompt the model to consider a particular country's perspective, responses shift to be more similar to the opinions of the prompted populations, but can reflect harmful cultural stereotypes. When we translate GlobalOpinionQA questions to a target language, the model's responses do not necessarily become the most similar to the opinions of speakers of those languages. We release our dataset for others to use and build on. Our data is at https://huggingface.co/datasets/Anthropic/llm_global_opinions. We also provide an interactive visualization at https://llmglobalvalues.anthropic.com.
Subfields: Artificial Intelligence, General Social Sciences
View Paper: https://www.semanticscholar.org/paper/534c58762e69d7afbcb0f6a7e53c07484f6d4891
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Lester James Validad Miranda
- Nathan Lambert
- Jacob Daniel Morrison
- David Q. Sun
- Hadas Kotek
- Khyathi Raghavi Chandu
- Nouha Dziri
- Sachin Kumar
In General Social Sciences
Citations in the corpus
- Gender bias and stereotypes in Large Language Models (2023, background): “Disparities in contributions to such data, for example in articles, citations, and editors of Wikipedia, as well as in testing of model outputs more generally, have been widely noted [e.g. 2, 27, 40, 52, 92, and citations therein].”
- The Claude 3 Model Family: Opus, Sonnet, Haiku (background): “…to Reduce Harms [11], Capacity for Moral Self-Correction in Large Language Models [12], Towards Measuring the Representation of Subjective Global Opinions in Language Models [13], Frontier Threats Red Teaming for AI Safety [14], and our Responsible Scaling Policy [5] to address catastrophic risks).”
- RewardBench: Evaluating Reward Models for Language Modeling (2024, background): “Initial work is studying this question for LLMs broadly, such as measuring representation (Durmus et al., 2023; Ryan et al., 2024) or moral foundations of LMs (Abdulhai et al., 2023), but this work should be extended to reward models.”
- High-Dimension Human Value Representation in Large Language Models (2024, unknown): “To better understand the human values of LLMs, one can use surveys of human values to query LLMs [13, 14, 23, 24].”