telemarketing_sales_rule
- Task Description: Determine how 16 C.F.R. § 310.3(a)(1) and 16 C.F.R. § 310.3(a)(2) apply to different fact patterns.
- Task Type: Binary classification
- Document Type: fact pattern
- Number of Samples: 51
- Input Length Range: 57-142 tokens
- Evaluation Metrics: accuracy (maximize), balanced_accuracy (maximize), f1_macro (maximize), f1_micro (maximize), valid_predictions_ratio (maximize)
- Tags: consumer protection, rule application
- 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 | 0.894 | 0.894 | 0.890 | 0.894 | 1.000 | 2025-08-04 | View |
| 2 | meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo | 0.830 | 0.823 | 0.823 | 0.830 | 1.000 | 2025-07-25 | View |
| 3 | gpt-4o-mini | 0.766 | 0.744 | 0.750 | 0.766 | 1.000 | 2025-07-02 | View |
| 4 | gpt-4.1-nano | 0.745 | 0.777 | 0.745 | 0.745 | 1.000 | 2025-07-03 | View |
| 5 | claude-3-5-haiku-20241022 | 0.723 | 0.666 | 0.662 | 0.723 | 1.000 | 2025-08-01 | View |
| 6 | google/gemma-2-27b-it | 0.702 | 0.640 | 0.628 | 0.702 | 1.000 | 2025-07-24 | View |
| 7 | claude-3-haiku-20240307 | 0.681 | 0.631 | 0.626 | 0.681 | 1.000 | 2025-07-25 | View |