maud_knowledge_definition
- Task Description: The model must select the answer that best characterizes the concept of Knowledge in a merger agreement.
- Task Type: Binary classification
- Document Type: merger agreement
- Number of Samples: 168
- Input Length Range: 11-1691 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 | meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo | 0.563 | 0.534 | 0.433 | 0.563 | 1.000 | 2025-07-25 | View |
| 2 | claude-3-5-haiku-20241022 | 0.563 | 0.539 | 0.482 | 0.563 | 1.000 | 2025-08-01 | View |
| 3 | google/gemma-2-27b-it | 0.509 | 0.484 | 0.408 | 0.509 | 1.000 | 2025-07-24 | View |
| 4 | gpt-4.1-nano | 0.491 | 0.481 | 0.474 | 0.491 | 1.000 | 2025-07-03 | View |
| 5 | gpt-4o-mini | 0.485 | 0.511 | 0.406 | 0.485 | 1.000 | 2025-07-02 | View |
| 6 | claude-3-haiku-20240307 | 0.461 | 0.452 | 0.447 | 0.461 | 1.000 | 2025-07-25 | View |
| 7 | meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo | 0.461 | 0.467 | 0.459 | 0.461 | 1.000 | 2025-08-03 | View |