BeIR

/

query-gen-msmarco-t5-large-v1

like 13

Hi, thank you for sharing this model card!

I am wondering about the training script and training arguments used for this model card, such as:

• T5 initial checkpoint (Did you use google/t5-v1_1-large?),
• maximum source/target length,
• batch size,
• optimizer,
• learning rate,
• learning rate scheduling,
• warmup steps,
• and total training steps.

Did you use the same training script and arguments used in doc2query/msmarco-t5-base-v1?

Best regards,
Jihyuk Kim