Tatoeba baseline from XTREME-R

[pefimov]

It is not clear which model is used for Tatoeba task.

XTREME-R paper says:

To make the setting more realistic, we move away from zero-shot retrieval and fine-tune models on SQuAD v1.1.

Script run_tatoeba.sh says:
# Add fine-tuned model path here
MODEL=/mnt/disk-1/models/squad/xlm-roberta-large_LR3e-5_EPOCH2.0_maxlen384_batchsize2_gradacc16

Does it mean that authors use model fine-tuned on original SQuAD for MRC task? Or it is something similar to LAReQA baseline?