AlibabaResearch / AdvancedLiterateMachinery

A collection of original, innovative ideas and algorithms towards Advanced Literate Machinery. This project is maintained by the OCR Team in the Language Technology Lab, Tongyi Lab, Alibaba Group.
Apache License 2.0
1.33k stars 161 forks source link

In Token classification, Is geolayoutlm using both entity extraction and entity linking #51

Open ManikantaNT opened 11 months ago

ccx1997 commented 11 months ago

We use both tasks to finetune the model, but the evaluation on each task is independent.

ManikantaNT commented 11 months ago

Thanks for your reply, but my concerns is that I fintuned the model for token classification but in the training phase the model is handling both entity extraction and entity linking tasks at same time is there any dependency between them or both are independent tasks

ManikantaNT commented 11 months ago

Hi Alibaba Reasearch Team, Thanks for your response, But my question is, I used the "geloayoutlm" for token classification, and during the fine-tuning process, I observed that both entity extraction and entity linking were trained simultaneously. I'm wondering if these two tasks are interdependent or if they are separate and don't rely on each other.

Thanks, Manikanta Ruppa


From: Changxu Cheng @.> Sent: Monday, September 25, 2023 1:02 PM To: AlibabaResearch/AdvancedLiterateMachinery @.> Cc: Ruppa Manikanta @.>; Author @.> Subject: Re: [AlibabaResearch/AdvancedLiterateMachinery] In Token classification, Is geolayoutlm using both entity extraction and entity linking (Issue #51)

You don't often get email from @.*** Learn why this is importanthttps://aka.ms/LearnAboutSenderIdentification

Careful, External Email

We use both tasks to finetune the model, but the evaluation on each task is independent.

— Reply to this email directly, view it on GitHubhttps://github.com/AlibabaResearch/AdvancedLiterateMachinery/issues/51#issuecomment-1733081476, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A25Q5J7FBWYZI4P4LXFWPJ3X4EXR7ANCNFSM6AAAAAA5FB3B7E. You are receiving this because you authored the thread.Message ID: @.***>

Disclaimer: This e-mail and any attachment may contain confidential, proprietary or legally privileged information. If you are not the original intended recipient and have erroneously received this message, you are prohibited from using, copying, altering or disclosing the content of this message. Please delete it immediately and notify the sender. Newgen Software Technologies Ltd (NSTL) accepts no responsibilities for loss or damage arising from the use of the information transmitted by this email including damages from virus and further acknowledges that no binding nature of the message shall be implied or assumed unless the sender does so expressly with due authority of NSTL.

ccx1997 commented 11 months ago

Hi, Manikanta, The two tasks are separately performed by two heads respectively. We only conducted multi-task finetuning.

ManikantaNT commented 11 months ago

Thanks for confirmation!