Understanding the paper : Persistent Anti-Muslim Bias in Large Language Models
Contributors
- Objectives
- This will help users to understand that how large language models such as GPT-3 capture racial bias
- Potential Use Cases
- Debiasing the language model
- Who is This For ?
- INTERMEDIATENLP Data scientist from all audience levels
Click on each of the following annotated items to see details.
ARTICLE 1. Bias in Machine learning models
- What is bias and sources of bias in ML models?
10 minutes
ARTICLE 2. Algorithmic Bias in Natural language processing models
- What are type of biases and algorithmic bias?
30 minutes
PAPER 3. Understanding the GPT-3
- What is GPT-3 ?
- How is GPT-3 different from previous transformer based architectures?
- How GPT-3 uses Few shot learning and zero shot learning to eliminate fine-tuning and the need for large task specific datasets?
25 minutes
ARTICLE 4. Persistent Anti-Muslim Bias in Large Language Models
- What type of bias is found in GPT-3?
- How to debias GPT-3 by introducing positive words and phrases?
15 minutes
Concepts Covered
0 comment