Open-book Question Answering on OpenBookQA FS n=30
38.47AccuracyMTO-MCP
Evaluation results
| Method | Links | |
|---|---|---|
| MTO-MCPEvaluation Protocol=Fine-Tuning, Backbone=T5-large, Adaptation Objective=LM2026.06 | 38.47 | |
| MTO-APEvaluation Protocol=Fine-Tuning, Backbone=T5-large2026.06 | 32.73 | |
| MTO-PreMAPEvaluation Protocol=Prompt-Tuning, Backbone=T5-large, Adaptation Objective=LM2026.06 | 25.23 | |
| MTO-MPEvaluation Protocol=Fine-Tuning, Backbone=T5-large2026.06 | 23.93 | |
| T5-large (baseline)Evaluation Protocol=Fine-Tuning, Backbone=T5-large2026.06 | 23.73 | |
| MTO-PreAPEvaluation Protocol=Prompt-Tuning, Backbone=T5-large, Adaptation Objective=Denoising2026.06 | 4.2 |