krangelie/bias-in-german-nlg
Master thesis: Exploring bias in German NLG (GPT-3 & GerPT-2). Applies regard classification and bias mitigation triggers.
This project helps evaluate and reduce unwanted biases in German text generated by AI models like GPT-3 or GerPT-2. It takes generated German text and analyzes it for 'regard' (how positive or negative the sentiment is towards different groups). The output helps researchers and practitioners understand and mitigate stereotypes in AI-generated German language.
No commits in the last 6 months.
Use this if you are developing or using German language generation models and need to identify or reduce biases related to sentiment or social stereotypes.
Not ideal if you are working with languages other than German, or if your primary concern is not bias in text generation.
Stars
16
Forks
1
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Sep 25, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/krangelie/bias-in-german-nlg"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
dccuchile/wefe
WEFE: The Word Embeddings Fairness Evaluation Framework. WEFE is a framework that standardizes...
dreji18/Fairness-in-AI
Detecting Bias and ensuring Fairness in AI solutions
amazon-science/bold
Dataset associated with "BOLD: Dataset and Metrics for Measuring Biases in Open-Ended Language...
dhfbk/variationist
Variationist: Exploring Multifaceted Variation and Bias in Written Language Data (ACL 2024 demo track)
soarsmu/BiasFinder
BiasFinder | IEEE TSE | Metamorphic Test Generation to Uncover Bias for Sentiment Analysis Systems