Instruction Hierarchy Robustness on RealGuardrails Distractors
0.95ScoreGPT-5-Mini-R
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5-Mini-R2026.03 | 0.95 | |
| GPT-5-Mini2026.03 | 0.88 |
Loading the SOTA2 catalog…
| Method | Links | |
|---|---|---|
| GPT-5-Mini-R2026.03 | 0.95 | |
| GPT-5-Mini2026.03 | 0.88 |