ArvinZhuang / DSI-QG

The official repository for "Bridging the Gap Between Indexing and Retrieval for Differentiable Search Index with Query Generation", Shengyao Zhuang, Houxing Ren, Linjun Shou, Jian Pei, Ming Gong, Guido Zuccon and Daxin Jiang.
MIT License
110 stars 16 forks source link

Data Files for NQ #4

Open Gouzi3618 opened 1 year ago

Gouzi3618 commented 1 year ago

Hi @ArvinZhuang,

Thanks for your great efforts! I was wondering if it is possible to provide the data files for the NQ dataset? Thanks!

ArvinZhuang commented 1 year ago

Hi @Gouzi3618 , thanks for your kind words :)

You can take a look at our old repo, in which we have code for processing NQ dataset.

Gouzi3618 commented 1 year ago

Get it, thanks! I was wondering if the results are improved with the current DSI-QG code?

ArvinZhuang commented 1 year ago

yeah, I think so. I remember scores can be improved by just randomly assigning docids rather than assigning ids sequentially.