lingaccept/cola@ukp roberta-base
1 version Architecture: houlsby

Adapter (with head) trained using the `run_glue.py` script with an extension that retains the best checkpoint (out of 30 epochs).

lingaccept/cola@ukp roberta-large
1 version Architecture: pfeiffer

Adapter (with head) trained using the `run_glue.py` script with an extension that retains the best checkpoint (out of 30 epochs).

lingaccept/cola@ukp distilbert-base-uncased
1 version Architecture: houlsby

Adapter for distilbert-base-uncased in Houlsby architecture trained on the CoLA dataset for 15 epochs with early stopping and a learning rate of 1e-4.

lingaccept/cola@ukp bert-base-uncased
1 version Architecture: houlsby

Adapter in Houlsby architecture trained on the CoLA task for 20 epochs with early stopping and a learning rate of 1e-4. See https://arxiv.org/pdf/2007.07779.pdf.

lingaccept/cola@ukp roberta-large
1 version Architecture: houlsby

Adapter (with head) trained using the `run_glue.py` script with an extension that retains the best checkpoint (out of 30 epochs).

lingaccept/cola@ukp bert-base-uncased
1 version Architecture: pfeiffer

Adapter in Pfeiffer architecture trained on the CoLA task for 20 epochs with early stopping and a learning rate of 1e-4. See https://arxiv.org/pdf/2007.07779.pdf.

lingaccept/cola@ukp distilbert-base-uncased
1 version Architecture: pfeiffer

Adapter for distilbert-base-uncased in Pfeiffer architecture trained on the CoLA dataset for 15 epochs with early stopping and a learning rate of 1e-4.

lingaccept/cola@ukp roberta-base
1 version Architecture: pfeiffer

Adapter (with head) trained using the `run_glue.py` script with an extension that retains the best checkpoint (out of 30 epochs).