corporate_lobbying
- Task Description: The Corporate Lobbying task requires an LLM to determine whether a proposed Congressional bill may be relevant to a company based on a company’s self-description in its SEC 10K filing.
- Task Type: Binary classification
- Document Type: corporate disclosure
- Number of Samples: 500
- Input Length Range: 30-2880 tokens
- Evaluation Metrics: accuracy (maximize), balanced_accuracy (maximize), f1_macro (maximize), f1_micro (maximize), valid_predictions_ratio (maximize)
- Tags: corporate law, issue spotting
- 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 | google/gemma-2-27b-it | 0.820 | 0.791 | 0.787 | 0.820 | 1.000 | 2025-07-24 | View |
| 2 | gpt-4o-mini | 0.810 | 0.797 | 0.782 | 0.810 | 1.000 | 2025-07-02 | View |
| 3 | meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo | 0.794 | 0.784 | 0.766 | 0.794 | 1.000 | 2025-07-24 | View |
| 4 | gpt-4.1-nano | 0.780 | 0.674 | 0.691 | 0.780 | 1.000 | 2025-07-03 | View |
| 5 | meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo | 0.759 | 0.779 | 0.740 | 0.759 | 1.000 | 2025-07-29 | View |
| 6 | claude-3-5-haiku-20241022 | 0.729 | 0.723 | 0.700 | 0.729 | 1.000 | 2025-08-01 | View |
| 7 | claude-3-haiku-20240307 | 0.682 | 0.726 | 0.670 | 0.682 | 1.000 | 2025-07-25 | View |