Gender Bias Evaluation on RealWorldQuestioning Education Recommendations
2Shannon Entropy (T-stat)ChatGPT-4-turbo
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| ChatGPT-4-turboIteration=1, Evaluation Protocol=Female vs Male T-test2025.05 | 2 | 0.04 | -0.87 | 0.38 | 0.44 | 0.66 | |
| Llama-3Iteration=1, Evaluation Protocol=Female vs Male T-test2025.05 | -0.24 | 0.8 | 0.01 | 0.98 | -1.06 | 0.28 | |
| DeepSeek-R1Iteration=1, Evaluation Protocol=Female vs Male T-test2025.05 | -1.17 | 0.24 | 0.86 | 0.38 | -1.84 | 0.06 | |
| ChatGPT-3.5-turboIteration=1, Evaluation Protocol=Female vs Male T-test2025.05 | -1.44 | 0.15 | -0.89 | 0.36 | -0.52 | 0.59 |