legal_reasoning_causality
- Task Description: Classify whether a district court opinion excerpt relies on statistical evidence in its reasoning for finding causality.
- Task Type: Binary classification
- Document Type: judicial text
- Number of Samples: 59
- Input Length Range: 141-1018 tokens
- Evaluation Metrics: accuracy (maximize), balanced_accuracy (maximize), f1_macro (maximize), f1_micro (maximize), valid_predictions_ratio (maximize)
- Tags: labor law, rhetorical understanding
- 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 | 0.909 | 0.910 | 0.908 | 0.909 | 1.000 | 2025-08-03 | View |
| 2 | gpt-4o-mini | 0.873 | 0.882 | 0.873 | 0.873 | 1.000 | 2025-07-02 | View |
| 3 | meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo | 0.873 | 0.882 | 0.873 | 0.873 | 1.000 | 2025-07-25 | View |
| 4 | google/gemma-2-27b-it | 0.867 | 0.864 | 0.864 | 0.867 | 0.545 | 2025-07-24 | View |
| 5 | claude-3-5-haiku-20241022 | 0.850 | 0.847 | 0.847 | 0.850 | 0.727 | 2025-08-01 | View |
| 6 | claude-3-haiku-20240307 | 0.667 | 0.750 | 0.667 | 0.667 | 0.055 | 2025-07-25 | View |
| 7 | gpt-4.1-nano | 0.564 | 0.589 | 0.556 | 0.564 | 1.000 | 2025-07-03 | View |