Medical QA on Medical
100Clean Accuracy (Eager)Llama-3.2-1B-Instruct
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Llama-3.2-1B-InstructExecution Backend=CUDAGraphs, Attack=Attack II2026.05 | 100 | 100 | 46.3 | 56.2 | |
| Llama-3.2-3B-InstructExecution Backend=CUDAGraphs, Attack=Attack II2026.05 | 100 | 100 | 51.2 | 55 | |
| Qwen2.5-3B-InstructExecution Backend=CUDAGraphs, Attack=Attack II2026.05 | 100 | 100 | 51.2 | 57.5 | |
| Qwen2.5-1.5B-InstructExecution Backend=CUDAGraphs, Attack=Attack II2026.05 | 55 | 52.5 | 56.2 | 65 |