Text Generation on Medical Chatbot
100ASRBaseline
Evaluation Results
| Method | Links | |
|---|---|---|
| BaselineModel=DialoGPT-small, Attack=Referral2025.12 | 100 | |
| BaselineModel=DialoGPT-medium, Attack=Referral2025.12 | 100 | |
| BaselineModel=Pythia-1B, Attack=Referral2025.12 | 90.34 | |
| BTUModel=DialoGPT-medium, Attack=Referral2025.12 | 89.18 | |
| STRIPModel=Pythia-1B, Attack=Referral2025.12 | 88.92 | |
| ONIONModel=DialoGPT-small, Attack=Referral2025.12 | 85.87 | |
| STRIPModel=DialoGPT-medium, Attack=Referral2025.12 | 85.72 | |
| MDPModel=DialoGPT-medium, Attack=Referral2025.12 | 82.56 | |
| ONIONModel=Pythia-1B, Attack=Referral2025.12 | 81.67 | |
| ONIONModel=DialoGPT-medium, Attack=Referral2025.12 | 78.29 | |
| BTUModel=GPT2-large, Attack=Referral2025.12 | 76.52 | |
| BaselineModel=OPT-125M, Attack=Referral2025.12 | 74.26 | |
| MDPModel=Pythia-1B, Attack=Referral2025.12 | 74.2 | |
| STRIPModel=DialoGPT-small, Attack=Referral2025.12 | 71.26 | |
| BaselineModel=GPT2-large, Attack=Referral2025.12 | 70.25 | |
| ONIONModel=OPT-125M, Attack=Referral2025.12 | 68.93 | |
| BTUModel=DialoGPT-small, Attack=Referral2025.12 | 68.75 | |
| BTUModel=Pythia-1B, Attack=Referral2025.12 | 68.74 | |
| BaselineModel=SmolLM2-135M-Instruct, Attack=Referral2025.12 | 68.44 | |
| ONIONModel=SmolLM2-135M-Instruct, Attack=Referral2025.12 | 62.78 | |
| BTUModel=OPT-125M, Attack=Referral2025.12 | 61.29 | |
| BTUModel=SmolLM2-135M-Instruct, Attack=Referral2025.12 | 59.67 | |
| STRIPModel=SmolLM2-135M-Instruct, Attack=Referral2025.12 | 59.15 | |
| MDPModel=DialoGPT-small, Attack=Referral2025.12 | 58.43 | |
| ONIONModel=GPT2-large, Attack=Referral2025.12 | 54.73 | |
| STRIPModel=OPT-125M, Attack=Referral2025.12 | 54.67 | |
| SCOUTModel=Pythia-1B, Attack=Referral2025.12 | 52.28 | |
| STRIPModel=GPT2-large, Attack=Referral2025.12 | 47.38 | |
| MDPModel=SmolLM2-135M-Instruct, Attack=Referral2025.12 | 46.92 | |
| SCOUTModel=SmolLM2-135M-Instruct, Attack=Referral2025.12 | 41.23 | |
| MDPModel=OPT-125M, Attack=Referral2025.12 | 39.84 | |
| SCOUTModel=GPT2-large, Attack=Referral2025.12 | 34.67 | |
| SCOUTModel=OPT-125M, Attack=Referral2025.12 | 34.56 | |
| MDPModel=GPT2-large, Attack=Referral2025.12 | 32.91 | |
| BaselineModel=GPT2-medium, Attack=Referral2025.12 | 30.67 | |
| STRIPModel=GPT2-medium, Attack=Referral2025.12 | 23.89 | |
| SCOUTModel=DialoGPT-medium, Attack=Referral2025.12 | 23.45 | |
| SCOUTModel=DialoGPT-small, Attack=Referral2025.12 | 22.91 | |
| BTUModel=GPT2-medium, Attack=Referral2025.12 | 19.76 | |
| MDPModel=GPT2-medium, Attack=Referral2025.12 | 11.32 | |
| ONIONModel=GPT2-medium, Attack=Referral2025.12 | 0 | |
| SCOUTModel=GPT2-medium, Attack=Referral2025.12 | 0 |