Counterfactual Statute Prediction on Legal Statute Dataset (Counterfactual)
81.01Overall AccuracyLexGuard
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| LexGuardBase Model=GPT-5.22026.05 | 81.01 | 83.53 | 77.02 | 2.42 | |
| LLM-J&H-Few-shotBase Model=GPT-5.2, Prompting Strategy=Few-shot2026.05 | 73.6 | 73.59 | 73.61 | 9.2 | |
| LLM-J&H-CoTBase Model=GPT-5.2, Prompting Strategy=Chain-of-Thought2026.05 | 73.41 | 75.03 | 70.76 | 9.96 | |
| LLMBase Model=GPT-5.22026.05 | 72.35 | 74.34 | 69.11 | 9.04 |