Open-book Question Answering on OpenBookQA 5% n=298
53.6AccuracyMTO-MP
Evaluation results
| Method | Links | |
|---|---|---|
| MTO-MPEvaluation Protocol=Fine-Tuning, Backbone=T5-large2026.06 | 53.6 | |
| MTO-APEvaluation Protocol=Fine-Tuning, Backbone=T5-large2026.06 | 52.2 | |
| MTO-MCPEvaluation Protocol=Fine-Tuning, Backbone=T5-large2026.06 | 50.68 | |
| T5-large (baseline)Evaluation Protocol=Fine-Tuning, Backbone=T5-large2026.06 | 49.21 | |
| MVP-Tuning2026.06 | 39.6 | |
| GreaseLM2026.06 | 39 | |
| SAFE2026.06 | 38.8 | |
| MHGRN2026.06 | 38 | |
| RoBERTa-largeBackbone=RoBERTa-large2026.06 | 37 | |
| MTO-PreMAPEvaluation Protocol=Prompt-Tuning, Backbone=T5-large, Adaptation Objective=LM2026.06 | 34.33 | |
| QA-GNN2026.06 | 33.53 | |
| MTO-PreAPEvaluation Protocol=Prompt-Tuning, Backbone=T5-large, Adaptation Objective=LM2026.06 | 32.71 | |
| GSC2026.06 | 29.6 |