Model Merging on 7 NLP tasks (test)
79.2AccuracyEXPERTS
Evaluation Results
| Method | Links | |
|---|---|---|
| EXPERTSModel=T5-L, Data-free=N/A, Fine-tuning Protocol=LoRA2026.04 | 79.2 | |
| REGMEANModel=T5-L, Data-free=false, Fine-tuning Protocol=LoRA2026.04 | 77.7 | |
| ACTMATModel=T5-L, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 76.7 | |
| EXPERTSModel=T5-B, Data-free=N/A, Fine-tuning Protocol=LoRA2026.04 | 76.5 | |
| REGMEANModel=T5-B, Data-free=false, Fine-tuning Protocol=LoRA2026.04 | 75.9 | |
| TSVModel=T5-L, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 75.4 | |
| TSVModel=T5-B, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 72.2 | |
| ACTMATModel=T5-B, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 72 | |
| K-TSVModel=T5-L, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 71.1 | |
| TAModel=T5-L, Data-free=false, Fine-tuning Protocol=LoRA2026.04 | 70.9 | |
| K-ISO-CModel=T5-L, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 69.4 | |
| K-TSVModel=T5-B, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 67.6 | |
| TAModel=T5-B, Data-free=false, Fine-tuning Protocol=LoRA2026.04 | 66.7 | |
| K-ISO-CModel=T5-B, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 64.7 | |
| AVERAGEModel=T5-B, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 62.2 | |
| AVERAGEModel=T5-L, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 57.2 | |
| ISO-CModel=T5-B, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 56.1 | |
| TAModel=T5-L, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 55 | |
| TAModel=T5-B, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 54.2 | |
| ZEROSHOTModel=T5-B, Data-free=N/A, Fine-tuning Protocol=LoRA2026.04 | 54 | |
| ISO-CModel=T5-L, Data-free=true, Fine-tuning Protocol=LoRA2026.04 | 52.1 | |
| ZEROSHOTModel=T5-L, Data-free=N/A, Fine-tuning Protocol=LoRA2026.04 | 51.3 |