Summary of Strategic Demonstration Selection For Improved Fairness in Llm In-context Learning, by Jingyu Hu et al.
Strategic Demonstration Selection for Improved Fairness in LLM In-Context Learning
by Jingyu Hu, Weiru Liu, Mengnan Du
First submitted to arxiv on: 19 Aug 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computation and Language (cs.CL); Computers and Society (cs.CY)
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 paper investigates the fairness implications of using in-context learning (ICL) prompts to steer large language models (LLMs) for processing tabular data. The study finds that deliberately including minority group samples in prompts boosts fairness without sacrificing predictive accuracy, and that the proportion of minority to majority samples affects the trade-off between fairness and prediction accuracy. To mitigate these issues, a new technique is introduced that employs clustering and evolutionary strategies to curate a diverse and representative sample set from the training data. This approach aims to enhance both predictive performance and fairness in ICL applications. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study looks at how to make sure large language models are fair when they’re learning about structured data like tables. The researchers found that if you include examples of minority groups in the prompts, it makes the model more fair without hurting its ability to predict things correctly. They also discovered that changing the mix of minority and majority group examples affects how well the model balances fairness and accuracy. To solve this problem, they came up with a new way to prepare the training data using clustering and evolutionary strategies. This method tries to make the model both more accurate and fair. |
Keywords
» Artificial intelligence » Clustering
