Is this model only works for English texts. Long texts will be truncated to at most 512 tokens? [E5: Text Embeddings by Weakly-Supervised Contrastive Pre-training]

[EasyLuck]

Hi, This issue is about e5:Text Embeddings by Weakly-Supervised Contrastive Pre-training. Liang Wang, Nan Yang, Xiaolong Huang, Binxing Jiao, Linjun Yang, Daxin Jiang, Rangan Majumder, Furu Wei, arXiv 2022 I see the informations on hugginface. Limitations: This model only works for English texts. Long texts will be truncated to at most 512 tokens. How can I fit it to Chinese text (中文文本)？ Train the model with Chinese text? if that , what the hardware configuration can do it. Thank you. Looking forward your reply.

[intfloat]

Yes, currently it only works for English.

We'll release multilingual versions of text embeddings in the coming month (no guarantee about the timeline though), please stay tuned!

Thanks, Liang

[iamlockelightning]

So looking forward to the release of multilingual E5!! 👀

[scy-flower]

Do you support Chinese now