ARR needs your help! We received 5800+ submissions for the June cycle, but with our current capacity, we can handle only half of these submissions. We can't start the review process without significant help. (1/2)

Website semantic-textual-relatedness.github.io I'm attending #NAACL2024 virtually but some co-organisers are in person. :-) More details on our datasets are shared in our upcoming #ACL2024NLP findings paper. x.com/nedjmaou/statu… [See you in Thailand hopefully!] 2/

Nedjma will be presenting our @SemRel2024 shared task at SemEval! The dataset paper will be presented at @acl

And we received great system description papers as well! :-D Congrats to team AAdaM for the SemEval best task description paper!

Yay! And our task received the best paper honourable mention award! Thanks Nedjma Ousidhoum نجمة أوسيدهم for leading and for all the team members.

The student ruined the semester for >200 students, the TAs, and myself. They will probably be graduating this July with 0 consequences. As educators, what kind of message are we sending the students? 2/

Sad to see such a behavior! I really hope that the police takes this issue seriously. I can't imagine what you've gone through Nedjma Ousidhoum نجمة أوسيدهم

👋🏻made it to Bangkok to attend #ACL2024NLP. We'll be presenting our paper SemRel2024: a collection of semantic textual relatedness datasets for 13 languages arxiv.org/pdf/2402.08638 (a team effort) at the in-person Findings poster session on Tue (12:15 pm) and at #Repl4NLP! 1/

Hi #NLProc-ers, if you work(ed) on mid/low-resource lgges [annotation/model construction/data collection,...], we're still looking for responses to this survey: forms.gle/qwi4jd5Av1xSvh… It takes a few minutes & many questions are optional. Deadline Aug 31, RTs appreciated!

☀️ We hope everyone had a great summer!!!☀️ Our 📢 NoDaLiDa/Baltic-HLT 2025 *2nd call for papers* is out! 📢 Deadline: 21 October ✍️⌚️ #NLProc #nodalida_baltichlt

We are glad to announce SemEval shared task 11: Bridging the Gap in Text-Based Emotion Detection. New textual datasets will be provided for 33 lgges (with many low-resource lgges)! Task GitHub: github.com/emotion-analys… CodaBench: codabench.org/competitions/3… Tracks👇🏻🧵1/

Hi #NLProc Twitter, we're trying to collect a few more responses so we'd be grateful if you could fill in this form forms.gle/8ape744yYseHzw… if you work/ed on NLP for low-resource languages. (Many questions are optional, deadline Oct 1st) RTs appreciated. Thank you!

The evaluation period of our Text Based Emotion Analysis (SemEval 2025 Task) SemEval shared task starts tomorrow! We have >30 new datasets including Algerian Arabic. (Dz Ar data is my translation of "La grande maison" by Mohammed Dib en.wikipedia.org/wiki/La_Grande…). Check out github.com/emotion-analys… for more details.

Check out AfriHate: a new collection of hate speech and abusive language datasets in 15 African languages (collected and annotated by native speakers) now accepted at #NAACL2025 #NLProc preprint: arxiv.org/abs/2501.08284 Big thanks to Shamsuddeen Hassan Muhammad, PhD for leading the project!

Co-led w\ Shamsuddeen Hassan Muhammad, PhD and Saif M. Mohammad Thanks to the great team members including Jan Philip Wahle terryruas Nirmal Surange Meriem Beloucif Christine de Kock Daniela Teodorescu and many others! Part of our Text Based Emotion Analysis (SemEval 2025 Task) SemEval task (Task paper: arxiv.org/abs/2503.07269) 2/2

Super happy and proud of this work on emotion for multiple languages! BRIGHTER will be at ACL 2025

I am so delighted to share the paper we have been working on with Nedjma Ousidhoum نجمة أوسيدهم and Saif M. Mohammad on multiple aspects to consider for low-resource data creation. Building Better will be presented at ACL 2025.

Optimization of LLMs without backpropagation? Check Alessio's (w/Zhenbang Dai, Katie Winkle and myself) work on Zeroth-Order optimization methods applied to NLP tasks such MT. The paper will presented in Vienna at @acl

🚨#ACL2025NLP position paper on key challenges in creating resources for mid- to low-resource lgges emphasising that data in #NLProc isn't merely a collection of tokens, but is rooted in cultural and social contexts. Based on 81 responses from researchers working on >70lgges 1/🧵

Our #NLProc #ACL2025 paper "Building Better: Avoiding Pitfalls in Developing Language Resources When Data Is Scarce" will be presented as an oral tomorrow (Wed) from 09:00–10:30 in Hall N2, and will be part of the multilinguality panel discussion moderated by thamar |. 1/2