stanford-futuredata / ColBERT

ColBERT: state-of-the-art neural search (SIGIR'20, TACL'21, NeurIPS'21, NAACL'22, CIKM'22, ACL'23, EMNLP'23)
MIT License
2.67k stars 355 forks source link

What is Colbert v1.9? #305

Open jordane95 opened 4 months ago

jordane95 commented 4 months ago

Hi, I'm a little confused about the version. Is this an intermediate checkpoint? How is it trained?

What is its difference with respect to v1 and v2?

Is training data for Colbert v1 available? Because the data downloaded from official site of msmarco doesn't fit into the format required by Trainer.