Argument Classification on Args.me (Accuracy)
92.4AccuracyGPT-5.2
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5.2Prompting/Aggregation Strategy=voting2026.02 | 92.4 | |
| GPT-5.2Prompting/Aggregation Strategy=mRaR2026.02 | 91.9 | |
| GPT-5.22026.02 | 91.9 | |
| gpt-oss-120bPrompting/Aggregation Strategy=voting2026.02 | 89.6 | |
| Pietron et al., 20242026.02 | 89.6 | |
| gpt-oss-120bPrompting/Aggregation Strategy=mRaR2026.02 | 88.5 | |
| gpt-oss-120b2026.02 | 88.5 | |
| gpt-oss-20bPrompting/Aggregation Strategy=voting2026.02 | 88.2 | |
| ds70bPrompting/Aggregation Strategy=voting2026.02 | 87 | |
| Llama-3.3-70BPrompting/Aggregation Strategy=voting2026.02 | 86.3 | |
| gpt-oss-20bPrompting/Aggregation Strategy=mRaR2026.02 | 86.2 | |
| gpt-oss-20b2026.02 | 86.2 | |
| Llama-3.1-70BPrompting/Aggregation Strategy=voting2026.02 | 85.7 | |
| BERT2026.02 | 85.3 | |
| Llama-4-Scout-70BPrompting/Aggregation Strategy=voting2026.02 | 85.2 | |
| ds70bPrompting/Aggregation Strategy=mRaR2026.02 | 84.4 | |
| ds70b2026.02 | 84.4 | |
| Llama-3.3-70BPrompting/Aggregation Strategy=mRaR2026.02 | 80.8 | |
| Llama 3.3 70b2026.02 | 80.8 | |
| Llama-3.1-70BPrompting/Aggregation Strategy=mRaR2026.02 | 80 | |
| Llama-4-Scout-70BPrompting/Aggregation Strategy=mRaR2026.02 | 79.6 | |
| Llama-3.3-70BPrompting/Aggregation Strategy=CoT2026.02 | 76.6 | |
| Llama-3.1-70BPrompting/Aggregation Strategy=CoT2026.02 | 75.8 | |
| Akiki and Potthast, 20202026.02 | 75.5 | |
| Llama-4-Scout-70BPrompting/Aggregation Strategy=CoT2026.02 | 74.2 | |
| Llama-3.1-8BPrompting/Aggregation Strategy=voting2026.02 | 73.2 | |
| Llama-3.1-8BPrompting/Aggregation Strategy=mRaR2026.02 | 69.6 | |
| Llama 3.1 8b2026.02 | 69.6 | |
| Llama-3.1-8BPrompting/Aggregation Strategy=CoT2026.02 | 65.2 | |
| Llama-3.1-3BPrompting/Aggregation Strategy=voting2026.02 | 61.4 | |
| Llama-3.1-3BPrompting/Aggregation Strategy=mRaR2026.02 | 58.8 | |
| Llama-3.1-3BPrompting/Aggregation Strategy=CoT2026.02 | 55.8 | |
| Llama-3.1-1BPrompting/Aggregation Strategy=mRaR2026.02 | 41.6 | |
| Llama-3.1-1BPrompting/Aggregation Strategy=voting2026.02 | 41.5 | |
| Llama-3.1-1BPrompting/Aggregation Strategy=CoT2026.02 | 41 |