maud_includes_consistent_with_past_practice
- Task Description: The model must determine if the Efforts Covenant clause in a merger agreement includes the phrase 'consistent with past practice'.
- Task Type: Binary classification
- Document Type: merger agreement
- Number of Samples: 182
- Input Length Range: 33-591 tokens
- Evaluation Metrics: accuracy (maximize), balanced_accuracy (maximize), f1_macro (maximize), f1_micro (maximize), valid_predictions_ratio (maximize)
- Tags: corporate law, interpretation
- 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 | gpt-4o-mini | 0.939 | 0.946 | 0.934 | 0.939 | 1.000 | 2025-07-02 | View |
| 2 | google/gemma-2-27b-it | 0.934 | 0.938 | 0.927 | 0.934 | 1.000 | 2025-07-24 | View |
| 3 | meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo | 0.928 | 0.930 | 0.921 | 0.928 | 1.000 | 2025-07-25 | View |
| 4 | meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo | 0.923 | 0.930 | 0.916 | 0.923 | 1.000 | 2025-08-03 | View |
| 5 | claude-3-5-haiku-20241022 | 0.757 | 0.813 | 0.755 | 0.757 | 1.000 | 2025-08-01 | View |
| 6 | claude-3-haiku-20240307 | 0.713 | 0.610 | 0.610 | 0.713 | 1.000 | 2025-07-25 | View |
| 7 | gpt-4.1-nano | 0.652 | 0.685 | 0.647 | 0.652 | 1.000 | 2025-07-03 | View |