:::information
Authors:
(1) Liang Wang, Microsoft Corporation, and correspondence to (Wangliang@microsoft.com);
(2) Nan Yang, Microsoft Company, and correspondence to (nanya@microsoft.com);
(3) Xiaolong Huang, Microsoft Corporation;
(4) Linjun Yang, Microsoft Corporation;
(5) Rangan Magmand, Microsoft Company;
(6) Furu Wei, Microsoft Corporation and Compreaseep to (Fuwei@microsoft.com).
:::
Links table
Abstract and 1 introduction
2 related work
3 method
3.1 Artificial data generation
3.2 Training
4 experiments
4.1 Artificial data statistics
4.2 Model model evaluation and evaluation
4.3 The main results
4.4 multi -language retrieval
5 analysis
5.1 Is training before training necessary?
5.2 extends to long text content and 5.3 analysis of excessive training
6 conclusion and references
Implementation details
The pollution group testing the test set
C is demanding the generation of artificial data
D instructions for training and evaluation
3.2 Training
\
::: Information about this paper Available on Arxiv Under CC0 1.0 Action License.
:::
\
