Natural Language Inference on QNLI (val)
88.05AccuracyDistilBERT
Evaluation results
| Method | Links | |
|---|---|---|
| DistilBERTPLM=Gold Data, Setting=SUPERVISED, #Param=66M2022.02 | 88.05 | |
| ZeroGenPLM=GPT2-XL, TAM=DistilBERT, Setting=ZEROGEN, #Param=66M2022.02 | 71.19 | |
| ZeroGenPLM=GPT2-Large, TAM=DistilBERT, Setting=ZEROGEN, #Param=66M2022.02 | 69.32 | |
| LSTMPLM=Gold Data, Setting=SUPERVISED, #Param=~7M2022.02 | 69 | |
| GPT2-XLSetting=PROMPTING, #Param=1.5B2022.02 | 60.6 | |
| ZeroGenPLM=GPT2, TAM=DistilBERT, Setting=ZEROGEN, #Param=66M2022.02 | 55.32 | |
| GPT2-LargeSetting=PROMPTING, #Param=762M2022.02 | 55.1 | |
| ZeroGenPLM=GPT2-XL, TAM=LSTM, Setting=ZEROGEN, #Param=~7M2022.02 | 52.26 | |
| ZeroGenPLM=GPT2-Large, TAM=LSTM, Setting=ZEROGEN, #Param=~7M2022.02 | 51.27 | |
| ZeroGenPLM=GPT2, TAM=LSTM, Setting=ZEROGEN, #Param=~7M2022.02 | 51.03 | |
| GPT2Setting=PROMPTING, #Param=117M2022.02 | 50.6 |