Kang Min Yoo

For those having trouble running the evaluation script in 2022 due to version updates in all the dependencies: use the following `requirements.txt` to get the specific package versions in the...

Could you elaborate more on the extrinsic evaluation?

3

You mentioned in the paper that you randomly sampled 1% of the training set and 5 of each class for the validation set. I tried to replicate the baseline results...

Missing paper in "Meta-applications - Dataset Construction"

1

Hi, thank you for working on a comprehensive survey of prompt-based methods. We would like to suggest the following paper be included under the corresponding topic section. **GPT3Mix: Leveraging Large-scale...

Working Barebone Example

2

In response to those who have trouble getting the script to work, I will soon include a simple working example to complement lacking documentation.

Can we get ballpark performance for the smallest model?

1

It would be great if you could share with us some rough results on SQuAD v2.0/v1.1 using the smallest model (117M). Thanks in advance!

The SST-2 dataset included in the repo contains 6,228 training samples, 692 validation samples, and 1821 test samples. But the official SST-2 dataset (which can be access via `torchtext`) contains...