privacy_policy_qa
- Task Description: Determine if a clause from a privacy policy contains enough information to answer a given question.
- Task Type: Binary classification
- Document Type: privacy policy
- Number of Samples: 10931
- Input Length Range: 11-240 tokens
- Evaluation Metrics: accuracy (maximize), balanced_accuracy (maximize), f1_macro (maximize), f1_micro (maximize), valid_predictions_ratio (maximize)
- Tags: interpretation, privacy law
- Paper: LegalBench: A Collaboratively Built Benchmark for Measuring Legal Reasoning in Large Language Models
- Dataset Download: https://hazyresearch.stanford.edu/legalbench/
7 submissions
| Rank | Model | accuracy | balanced_accuracy | f1_macro | f1_micro | valid_predictions_ratio | Date | Results |
|---|---|---|---|---|---|---|---|---|
| 1 | meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo | 0.720 | 0.720 | 0.720 | 0.720 | 1.000 | 2025-08-04 | View |
| 2 | google/gemma-2-27b-it | 0.700 | 0.700 | 0.698 | 0.700 | 1.000 | 2025-07-24 | View |
| 3 | meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo | 0.685 | 0.685 | 0.685 | 0.685 | 1.000 | 2025-07-25 | View |
| 4 | claude-3-5-haiku-20241022 | 0.651 | 0.646 | 0.644 | 0.651 | 0.585 | 2025-08-01 | View |
| 5 | gpt-4o-mini | 0.646 | 0.645 | 0.611 | 0.646 | 1.000 | 2025-07-02 | View |
| 6 | gpt-4.1-nano | 0.536 | 0.535 | 0.417 | 0.536 | 1.000 | 2025-07-03 | View |
| 7 | claude-3-haiku-20240307 | 0.530 | 0.532 | 0.481 | 0.530 | 0.989 | 2025-07-25 | View |