Non-existent links in some papers

[Kareem Darwish]

AA All,

               I was going through a few Arabic NLP papers today and the datasets mentioned therein sounded very interesting.  For the first 3 papers, I clicked on the links for the datasets and they were either non-existent links or empty datasets on Huggingface (see screenshot).  I was either having a really unlucky streak or this phenomenon is prevalent.  Either way, I behoove folks to actually put their data online if they claim in their papers that they are releasing the data.  You don’t have to release data if you don’t want to or you can’t.  However, if you say you are releasing the data and you provide a link, make sure that the data is public and the link works.

Kareem

 

[Abdul-Rahman Mawlood-Yunis]

Hi Kareem, 

Hope you are doing well. Typically  conferences require authors to hide their identities on submission. Providing links to datasets or other resources can compromise the anonymity of a blind review,

Regards,

AR

--
You received this message because you are subscribed to the Google Groups "SIGARAB: Special Interest Group on Arabic Natural Language Processing" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sigarab+unsubscribe@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sigarab/SA1PR20MB5502EC101298410374201AF8C4BEA%40SA1PR20MB5502.namprd20.prod.outlook.com.

[Kareem Darwish]

That is true. However, I am specifically talking about arxiv papers that have seemingly valid links to huggingface.co or github.com, but the links are not working. 

Get Outlook for Android

---

From: Abdul-Rahman Mawlood-Yunis <trustedpropert...@gmail.com>
Sent: Tuesday, December 30, 2025 7:49:13 AM
To: Kareem Darwish <kareem...@live.com>
Cc: sig...@googlegroups.com <sig...@googlegroups.com>
Subject: Re: [SIGARAB] Non-existent links in some papers

 

[Mahmoud Fawzi]

Hi Abdul-Rahman,

This is the case during the review process. However, after the paper gets accepted, authors should add the working links to the camera-ready version of the paper.

Regards

Mahmoud

On Tue, 30 Dec 2025 at 06:26, Abdul-Rahman Mawlood-Yunis <trustedpropert...@gmail.com> wrote:

Hi Kareem, 

Hope you are doing well. Typically  conferences require authors to hide their identities on submission. Providing links to datasets or other resources can compromise the anonymity of a blind review,

Regards,

AR

On Sunday, December 28, 2025, Kareem Darwish <kareem...@live.com> wrote:

AA All,

               I was going through a few Arabic NLP papers today and the datasets mentioned therein sounded very interesting.  For the first 3 papers, I clicked on the links for the datasets and they were either non-existent links or empty datasets on Huggingface (see screenshot).  I was either having a really unlucky streak or this phenomenon is prevalent.  Either way, I behoove folks to actually put their data online if they claim in their papers that they are releasing the data.  You don’t have to release data if you don’t want to or you can’t.  However, if you say you are releasing the data and you provide a link, make sure that the data is public and the link works.

Kareem

 

--
You received this message because you are subscribed to the Google Groups "SIGARAB: Special Interest Group on Arabic Natural Language Processing" group.

To unsubscribe from this group and stop receiving emails from it, send an email to sigarab+u...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/sigarab/SA1PR20MB5502EC101298410374201AF8C4BEA%40SA1PR20MB5502.namprd20.prod.outlook.com.

--

You received this message because you are subscribed to the Google Groups "SIGARAB: Special Interest Group on Arabic Natural Language Processing" group.

To unsubscribe from this group and stop receiving emails from it, send an email to sigarab+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sigarab/CAFW_6uPnxkwuFQ1he%3D6Tq9V8Q%2BAyOJ-meUmRO3H5fXKeejUedg%40mail.gmail.com.