maud_relational_language_(mae)_applies_to
- Task Description: The model must select the answer that best characterizes the carveouts pertaining to Material Adverse Effect (MAE) in a merger agreement excerpt.
- Task Type: 3-way classification
- Document Type: merger agreement
- Number of Samples: 91
- Input Length Range: 623-1744 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 | claude-3-haiku-20240307 | 0.778 | 0.524 | 0.482 | 0.778 | 1.000 | 2025-07-28 | View |
| 2 | gpt-4.1-nano | 0.700 | 0.523 | 0.521 | 0.700 | 1.000 | 2025-07-03 | View |
| 3 | meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo | 0.700 | 0.473 | 0.445 | 0.700 | 1.000 | 2025-08-03 | View |
| 4 | meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo | 0.689 | 0.532 | 0.358 | 0.689 | 1.000 | 2025-07-25 | View |
| 5 | google/gemma-2-27b-it | 0.600 | 0.441 | 0.298 | 0.600 | 1.000 | 2025-07-24 | View |
| 6 | gpt-4o-mini | 0.567 | 0.486 | 0.479 | 0.567 | 1.000 | 2025-07-02 | View |
| 7 | claude-3-5-haiku-20241022 | 0.289 | 0.437 | 0.287 | 0.289 | 1.000 | 2025-08-01 | View |