Generative Question Answering on SQuAD Double-Char (test)
69.52EMSDBN-p
Evaluation results
| Method | Links | |
|---|---|---|
| SDBN-pModel=Qwen-2.5-7B, PEFT Method=LoRA, Training Samples=2002026.06 | 69.52 | |
| EDAModel=Qwen-2.5-7B, PEFT Method=LoRA, Training Samples=2002026.06 | 50.32 | |
| FreeLBModel=Qwen-2.5-7B, PEFT Method=LoRA, Training Samples=2002026.06 | 49.68 | |
| NEFTuneModel=Qwen-2.5-7B, PEFT Method=LoRA, Training Samples=2002026.06 | 48.6 | |
| VanillaModel=Qwen-2.5-7B, PEFT Method=LoRA, Training Samples=2002026.06 | 48.28 |