Commonsense Question Answering on CommonSenseQA n=425 (5%)
61.22AccuracyMTO-AP
Evaluation Results
| Method | Links | |
|---|---|---|
| MTO-APEvaluation Protocol=Fine-Tuning, Backbone=T5-large, Adaptation Objective=Denoising2026.06 | 61.22 | |
| MTO-MCPEvaluation Protocol=Fine-Tuning, Backbone=T5-large, Adaptation Objective=Denoising2026.06 | 60.69 | |
| MTO-PreMAPEvaluation Protocol=Prompt-Tuning, Backbone=T5-large, Adaptation Objective=Denoising2026.06 | 52.35 | |
| MVP-Tuning2026.06 | 48.99 | |
| SAFE2026.06 | 36.45 | |
| MTO-PreAPEvaluation Protocol=Prompt-Tuning, Backbone=T5-large, Adaptation Objective=LM2026.06 | 33.41 | |
| QA-GNN2026.06 | 32.95 | |
| GSC2026.06 | 31.02 | |
| RoBERTa-largeBackbone=RoBERTa-large2026.06 | 29.66 | |
| MHGRN2026.06 | 29.01 | |
| MTO-MPEvaluation Protocol=Fine-Tuning, Backbone=T5-large2026.06 | 25.53 | |
| T5-large (baseline)Evaluation Protocol=Fine-Tuning, Backbone=T5-large2026.06 | 23.12 | |
| GreaseLM2026.06 | 22.8 |