Arid Hasan retweeted
MMCultureQA Participant Webinar | SemEval-2027 Task 6
We invite researchers, students, and participating teams to join our upcoming online webinar for MMCultureQA: Multilingual and Multimodal Culturally Grounded Spoken and Visual Question Answering.
The webinar will provide an overview of the shared task and help participants get started with the datasets and evaluation platform.
📅 Date: Wednesday, November 25, 2026
🕓 Time: 4:00 PM Qatar Time (UTC+3)
💻 Platform: Microsoft Teams
What will we cover?
Overview of the Spoken and Textual Visual QA tracks
Multilingual datasets and recently released languages
Development evaluation and leaderboards
Starter code, baselines, and submission process
Q&A with the organizing team
🏆 The development evaluation phase is currently underway! Training, development, and dev-test datasets are available for multiple languages.
✅ Interested in participating?
If you haven't registered for the shared task yet, please register here: shorturl.at/QkLl3
We will share the Microsoft Teams joining link with registered participants before the webinar. Those already registered for the shared task do not need to register again.
📂 Datasets: huggingface.co/datasets/QCRI…
💻 Starter Code: github.com/mmcultureqa-semev…
🌐 Task Website: mmcultureqa-semeval27.github…
We look forward to meeting participating teams, discussing the shared task, and answering your questions.
Please share this announcement with colleagues and students interested in multilingual and multimodal AI.
W/
@shammur_absar @_aridhasan @NadirDurrani5 @dalvifahim ....
#SemEval2027 #MMCultureQA #MultimodalAI #MultilingualNLP #AudioLLM #VisualQuestionAnswering #SpeechAI #CulturallyGroundedAI
🤖 Made with AI
Arid Hasan retweeted
#NLProc
More that 200 teams have already registered, if you have not registered already, please consider doing so.
SemEval-2027 Task 6 - MMCultureQA: Leaderboards Are Live & Participant Webinar!
We are pleased to share two important updates about MMCultureQA: Multilingual and Multimodal Culturally Grounded Spoken and Visual Question Answering.
🏆 1. Leaderboards Are Now Live!
Participants can now submit their systems and evaluate their performance.
Currently, the leaderboards support English, Modern Standard Arabic (MSA), Egyptian Arabic, and Levantine Arabic. Datasets and leaderboards for additional languages will be available soon.
🔗 Evaluation platforms:
eval.qcri.org/web/challenges…
eval.qcri.org/web/challenges…
📅 2. MMCultureQA Participant Webinar
We invite all registered participants and interested researchers to join our upcoming webinar!
🗓 Date: Wednesday, November 25, 2026
🕓 Time: 4:00 PM Qatar Time (UTC+3)
💻 Platform: Microsoft Teams (joining link to follow)
We will provide an overview of the shared task, datasets, evaluation setup, and submission process, followed by a Q&A session.
Please mark your calendars!
✅ Haven't registered yet? We welcome new teams and encourage interested researchers and students to register and participate.
🔗 Register here: shorturl.at/QkLl3
🌐 Task website: mmcultureqa-semeval27.github…
We look forward to your submissions and participation in the webinar. Please share this announcement with colleagues, students, and research groups interested in multilingual and multimodal AI!
#SemEval2027 #MMCultureQA #MultimodalAI #MultilingualNLP #VisualQuestionAnswering #SpeechAI #AudioLLM #CulturallyGroundedAI
Arid Hasan retweeted
First CFP - The Third Workshop on Bangla Language Processing (BLP 2027)
Theme: Beyond Text: Bangla in the Era of Reasoning, Agents, and Multimodal Intelligence
Co-located with NAACL 2027, San Francisco, California, United States, June 2027
Workshop website: blp-workshop.github.io/
Apologies for cross-posting.
Foundation models are moving language technology beyond text generation toward reasoning, retrieval, agents, speech, and multimodal interaction. However, many of these advances remain concentrated in resource-rich languages. Despite a long history of NLP and speech research, Bangla remains underrepresented in frontier AI systems, particularly in reasoning, agentic applications, culturally grounded knowledge, speech-native models, multimodal understanding, and safety. Further challenges arise from dialectal, code-mixed, transliterated, and diaspora varieties, as well as related languages such as Assamese.
The Third Workshop on Bangla Language Processing (BLP 2027) aims to bring together researchers working in NLP, speech, multimodal AI, and human-centered computing to advance language technologies that can understand, reason, communicate, and act effectively in diverse Bangla settings. Building on BLP 2023 and BLP 2025, this edition focuses on linguistically inclusive, culturally grounded, trustworthy, multimodal, and accessible AI.
We invite original research papers presenting new methods, datasets, benchmarks, evaluations, applications, and analyses. Bangla is the primary focus, while comparative studies involving related or underrepresented languages are welcome when they offer insights transferable to Bangla and other low-resource settings. We particularly encourage research that addresses real-world linguistic diversity and practical constraints on data and computation.
Topics of Interest
We welcome papers on topics including, but not limited to:
Foundation Models: Bangla and multilingual large language models; pretraining, adaptation, and instruction tuning; synthetic data; efficient models; cross-lingual transfer; and robustness to dialects, code-mixing, and transliteration.
Reasoning and Agents: Verifiable, symbolic, and neuro-symbolic reasoning; retrieval-augmented generation; long-context understanding and memory; tool use, planning, and code generation; and single- and multi-agent systems.
Speech and Conversational AI: Automatic speech recognition, speech synthesis, and AudioLLMs; spoken question answering and dialogue; code-switched speech; speaker and language recognition; and conversational systems.
Multimodal and Grounded AI: Vision-language and audio-visual models; multimodal question answering and reasoning; document and scene understanding; multimodal generation; and culturally grounded multimodal evaluation.
Trustworthy and Human-Centered AI: Safety and alignment; adversarial robustness; hallucination and factuality; bias, fairness, and interpretability; cultural understanding; human-AI interaction; accessibility; and responsible deployment.
Resources, Evaluation, and Efficient NLP: Corpora, datasets, and benchmarks; representative resources for underrepresented varieties; human and LLM-based evaluation; data-efficient learning and inference; and core NLP tasks such as machine translation, information extraction, question answering, and summarization.
Planned Shared Tasks
BLP 2027 also plans to organize three shared tasks:
Task 1 - Adversarial Prompt Classification: Identifying benign and adversarial prompts in Bangla, including jailbreaks, prompt injection, and code-mixed or transliterated attempts to bypass LLM safety safeguards.
Task 2 - Multilingual Question Answering: Culturally and regionally grounded open-ended QA in Bangla, English, Assamese, and Hindi, with monolingual QA and multilingual QA settings based on the MultiNativQA resource.
Task 3 - BLP Datathon: Collecting regional Bangla speech recordings and identifying where open-source automatic speech recognition (ASR) models struggle with different dialects and accents through community-driven participation.
Detailed task guidelines, datasets, evaluation procedures, and participation timelines will be announced separately at blp-workshop.github.io/share….
Important Dates
Paper submission deadline: February 15 (Monday), 2027
Notification of acceptance: March 26 (Friday), 2027
Camera-ready deadline: April 16 (Friday), 2027
Workshop: June 2027, co-located with NAACL 2027 (exact date TBA)
All deadlines are 11:59 p.m. Anywhere on Earth (AoE).
Submission Details
We welcome original, unpublished research, including completed work and promising work in progress. Submissions will undergo double-blind peer review, with acceptance decisions made by the workshop program chairs.
Submission categories are:
Full papers: Up to 8 pages.
Short papers: Up to 4 pages, including work in progress and innovative research ideas.
Shared-task system papers: Up to 4 pages.
Camera-ready versions may use up to 9 pages for full papers and 5 pages for short papers, subject to the final proceedings guidelines. Accepted papers will be presented orally or as posters.
All submissions must follow the official ACL/ARR two-column formatting guidelines. Authors should use the official style files without modification. We recommend checking submissions with ACL PubCheck, which should also be used for camera-ready papers. For double-blind reviewing and anonymity requirements, please consult the ARR anonymity guidelines.
Further details on submission procedures, including any ARR-reviewed paper route, will be announced on the workshop website.
Submission portal: TBA
Invitation to Review
Researchers with relevant publication experience who are interested in joining the program committee are invited to complete this reviewer interest form: forms.gle/zZiZvDQRjALHcGfG7
Mentoring Program
We offer a mentoring program to support students and newcomers. It includes pre-submission mentoring to help finalize paper drafts and a buddy program during the in-person workshop. If you would like to receive mentoring, please sign up using the mentoring registration form.
Sponsors
We invite companies and institutions to partner with us in advancing Bangla language technology and research. Sponsoring BLP 2027 provides an opportunity to engage with leading researchers, industry professionals, and emerging scholars from around the world while supporting innovation and collaboration in Bangla language processing.
Workshop Organizers
- Firoj Alam, Qatar Computing Research Institute, Qatar
- Sudipta Kar, Oracle Health AI, Seattle, USA
- Shammur Absar Chowdhury, Qatar Computing Research Institute, Qatar
- Nishat Raihan, University of Notre Dame, USA
- Md Arid Hasan, University of Toronto, Canada
- Naeemul Hassan, University of Maryland, USA
- Enamul Hoque Prince, York University, Canada
- Tasnim Mohiuddin, Qatar Computing Research Institute, Qatar
- Md Tahmid Rahman Laskar, Dialpad, Canada
W/
@cryptexcode, @shammur_absar, @_aridhasan, Nishat, @Enamul_Hoque, @tahmedge Naeemul Hassan, Tasnim Mohiuddin
🤖 Made with AI
Arid Hasan retweeted
📣 Call for participation: BLP-2027 Shared Task 2 on Multilingual Cultural Question Answering
Can your model answer the everyday questions people actually ask, in their own language and with local knowledge?
We're excited to announce Multilingual Cultural QA, a shared task at the Third Workshop on Bangla Language Processing (BLP), co-located with NAACL 2027. Systems answer culturally grounded questions from native speakers in Dhaka, Kolkata, Assam, Delhi, and Doha, in the same language they were asked: বাংলা · English · অসমীয়া · हिन्दी
🔗 Website, data, and code: multilingual-qa.github.io/
🏆 Submit your system: eval.qcri.org/web/challenges…
🧩 Two subtasks
- 2A, Monolingual QA: one model per language
- 2B, Multilingual QA: one model for all four languages, including low-resource Assamese
🎯 Eight tracks, so join in any way you like: one language, several, or all of them, in either or both subtasks. Every track has its own leaderboard.
📊 Data and evaluation
~37K training QA pairs from MultiNativQA, covering 18 everyday topics such as food, travel, religion, tradition, and education.
Ranked by BERTScore F1 with language-specific encoders; ROUGE, METEOR, and BLEU are also reported.
📊 Baseline
An mT5 baseline, a fine-tuning starter script, and the official scorer are provided
🗓️ Key dates
Development phase: open now
Test phase: Jan 15 – Feb 1, 2027
System papers due: Feb 15, 2027
✨ Two more shared tasks at BLP 2027:
Task 1, BanglaKnight (Organizer: Nishat Raihan, PhD): classify Bangla prompts, from harmless requests to jailbreak attempts, into four risk levels, across Bangla script, romanized Bangla, and code-mixed text.
Task 3, BLP Datathon (Organizer: Sudipta Kar): collect regional Bangla speech and find where open-source speech recognition struggles with different dialects and accents.
👉 All shared tasks: blp-workshop.github.io/share…
Students, researchers, and practitioners working on multilingual NLP, low-resource languages, or culturally aware LLMs are all welcome. Please share with anyone who might be interested! 🙌
w/ @_aridhasan @shammur_absar Sahinur Rahman Laskar @nowshining Syed Ishtiaque Ahmed
#NLProc #NAACL2027 #SharedTask #MultilingualNLP #LowResourceLanguages #QuestionAnswering #LLM #Bangla #Hindi #Assamese
Arid Hasan retweeted
📢 Call for Participation: SemEval-2027 Task 6: Multilingual and Multimodal Culturally Grounded Spoken and Visual Question Answering
We are pleased to invite participation in MMCultureQA, a SemEval-2027 Shared Task on Multilingual and Multimodal Culturally Grounded Question Answering.
🎯 What is MMCultureQA?
The goal of this shared task is to evaluate whether multimodal systems can answer questions that require not only understanding an image, but also linguistic and cultural knowledge.
Given an image and a spoken or written question, systems must generate a short, open-ended textual answer. Questions cover culturally grounded topics such as food, places, customs, everyday objects, traditions, and social practices, where visual information alone may not be sufficient.
The shared task includes two main tasks:
🎧 Task 1: Spoken Visual Question Answering
Input: Image + spoken question
Output: Open-ended textual answer
📝 Task 2: Textual Visual Question Answering
Input: Image + written question
Output: Open-ended textual answer
Each task includes multiple language tracks, and participants may take part in one or multiple tracks.
🌍 Languages include: English, Modern Standard Arabic, Dialectal Arabic (Egyptian, Leventine), Bangla, Hindi, Amharic, Assamese, Gujarati, Italian, Marathi, Oromo, Somali, Sinhala, Nepali, Tigrinya, Urdu.
📂 Dataset
The MMCultureQA data builds on the EverydayMMQA framework, which enables the development of culturally grounded multimodal QA datasets across image, text, and speech modalities.
📅 Important Dates
Training data release: 23 September 2026
Evaluation starts: January 10, 2027
Evaluation ends: January 31, 2027
System description papers: February 2027 (tentative)
Notification to authors: March 2027 (tentative)
Camera-ready papers: April 2027 (tentative)
SemEval-2027 Workshop: Summer 2027
🚀 How to Participate
✅ Register your team:
docs.google.com/forms/d/15-V…
💬 Join our Slack community for announcements, questions, and discussions:
join.slack.com/t/mm-eval/sha…
🌐 Task website:
mmcultureqa-semeval27.github…
📂 Dataset:
huggingface.co/datasets/QCRI…
💻 Scripts: github.com/mmcultureqa-semev…
We welcome participation from researchers and teams working on multimodal LLMs, vision-language models, AudioLLMs, multilingual NLP, speech processing, visual question answering, under-resourced languages, and culturally grounded AI.
Please feel free to share this call with colleagues, students, and research groups who may be interested.
We look forward to your participation in MMCultureQA at SemEval-2027!
- The MMCultureQA Organizing Team
#SemEval2027 #MultimodalAI #MultilingualNLP #NLproc
W/
@shammur_absar @_aridhasan @NadirDurrani5 @dalvifahim @UsmanNaseem87 @hunzed11 @Zein_5 @MKmainasi @ishtiaqueSIA @mahedmousavi @MdNishatRaihan, Syeda Faiza Ahmed, Sahinur Rahman Laskar, Marcos Zampieri, Giuseppe Riccardi, Vrunda N. Sukhadia, Ebru Arisoy, Ali Ezzat Shahroor
Excited to share that our research on global narrative dominance in LLMs was featured in @el_pais! The article explores how language and unequal online representation shape the narratives AI systems present as knowledge.
📰 elpais.com/tecnologia/2026-0…
#LLMs #ResponsibleAI #NLP
1
21
Arid Hasan retweeted
How well do today’s multimodal models understand Arabic language varieties and MENA culture, rather than simply recognizing objects or translating text?
ImageEval 2026 focuses on this gap by evaluating culturally grounded Arabic image understanding and generation. The shared task is part of ArabicNLP 2026, co-located with EMNLP 2026.
The two tracks are:
🔹 Culturally Grounded Arabic VQA and Hallucination Detection
🔹 Cultural Accuracy Evaluation for Text-to-Image Generation
Participants can submit system runs and write system-description papers for the ArabicNLP 2026 proceedings.
Registration:
- Spoken VQA (1a) English: codabench.org/competitions/1…
- Spoken VQA (1a) MSA: codabench.org/competitions/1…
- Hallucination (1b) English: codabench.org/competitions/1…
- Hallucination (1b) MSA: codabench.org/competitions/1…
- CRAI-Bench: codabench.org/competitions/1…
🌐 Website: imageeval2026.github.io/
💻 Data and baselines: github.com/ImageEval2026/Ima…
💬 Slack: join.slack.com/t/mm-eval/sha…
Extended deadlines
📅 August 2, 2026 — Registration deadline and start of the evaluation cycle; test sets released
📅 August 7, 2026 — End of the evaluation cycle; run submissions due
📅 August 9, 2026 — Leaderboard released
📅 August 19, 2026 — Shared-task papers due
📅 August 30, 2026 — Notification of acceptance
📅 September 10, 2026 — Camera-ready papers due
📅 October 24–29, 2026 — ArabicNLP 2026
W/
Samir Abdaljalil, Ahlam Bashiti, Farina Amir, @shammur_absar, @NadirDurrani5, @dalvifahim, Basel Mousi, Hunzalah Hassan Bhatti, Zien Sheikh Ali, Erchin Serpedin, Hasan Kurban, @mjarrar, @_aridhasan
@emnlpmeeting, @_ArabicNLP
#ImageEval2026 #ArabicNLP2026 #EMNLP2026 #MultimodalAI #ArabicNLP #ComputerVision #GenerativeAI #VisualQuestionAnswering #TextToImage #CulturalAI #SharedTask #NLProc
Arid Hasan retweeted
#NLProc,
Glad to share, nativeqa framework has received significant, over 2k downloads.
W/
@shammur_absar @_aridhasan ...
Arid Hasan retweeted
Welcome 2026–27 SRI Graduate Fellows!
We've welcomed fifteen @UofT researchers working across computer science, information studies, psychology, law, medicine, management, industrial relations, and history.
Meet our cohort & learn about their projects: uoft.me/crh
Arid Hasan retweeted
⏰ Just 2 weeks until the deadline! Submit your research to the 2nd Bangla Language Processing (BLP) Workshop, co-located with
@aaclmeeting
(AACL-IJCNLP 2025) in Mumbai. 📷 Deadline: Sept 29, 2025 📷blp-workshop.github.io #NLP #AACL
Arid Hasan retweeted
The Call for Papers for the second edition of the Bangla Language Processing (BLP) Workshop is out (blp-workshop.github.io/cfp).
This year's BLP workshop will be co-located with the IJCNLP-AACL conference from December 20-24, 2025, in Mumbai, India.
We invite researchers to submit original work presenting novel approaches and resources that advance Bangla and other low-resource languages.
This year's workshop will also consist of two shared tasks (blp-workshop.github.io/share…):
*Task 1: Hate Speech Detection (github.com/AridHasan/blp25_t…)
*Task 2: Code Generation (noshinulfat.github.io/blp25_…)
Important Dates:
*Paper submission due (both regular and shared tasks papers): Sep 29 (Mon), 2025
*Notification of acceptance: Nov 3 (Mon), 2025
*Camera-ready due: Nov 11 (Tue), 2025
@aaclmeeting @firojalam04 @shammur_absar @Enamul_Hoque @cryptexcode @MdNishatRaihan
#aacl #ijcnlp #nlproc #blp
Arid Hasan retweeted
#NLProc,
Are you looking to benchmark your LLMs for Modern Standard Arabic (MSA) and dialects? Look no further!
Check out AraDiCE! We’ve made the dataset and evaluation config publicly available, so you can seamlessly benchmark your LLMs for Arabic.
Curious to see which model is winning? Please take a look at the paper.
📜 Read the paper: arxiv.org/pdf/2409.11404
📂 Access the data: shorturl.at/Htgsu
🧪 Start evaluating: shorturl.at/yFvFR
Ready to get started? Open your terminal and give it a try! 🖥️💻
Attending @coling2025 ? You can learn more.
#ArabicNLP #LLMs #AraDiCE #Benchmarking #NaturalLanguageProcessing
W/ @baselmousi995 @NadirDurrani5 Fatema, @_aridhasan @MaramHasanain Tameem, @dalvifahim @shammur_absar
Arid Hasan retweeted
COLING 2025 Paper Acceptance!
Proud to share that our paper, "AraDiCE: Benchmarks for Dialectal and Cultural Capabilities in LLMs," has been accepted at COLING 2025! 🎉
If you're exploring Arabic LLMs and striving to enhance your models with dialectal and cultural capabilities, look no further—AraDiCE is the resource you've been looking for!
🔗 Read the paper: arxiv.org/abs/2409.11404
📂 Access the dataset: huggingface.co/collections/Q…
#NLProc, @coling2025 #ArabicLLM, #dialects #culturalunderstanding
W/ @baselmousi995 @NadirDurrani5 Fatema, @_aridhasan @MaramHasanain Tameem, @dalvifahim @shammur_absar
Arid Hasan retweeted
#NLProc, we have extended the submission deadline to 4th May, 2024 (AoE). You can consider submitting your system.
Submission System for the evaluation phase:
Task 1: codalab.lisn.upsaclay.fr/com…
Task 2: codalab.lisn.upsaclay.fr/com…
Please join on slack channel for your query: join.slack.com/t/araieval/sh…
@aclmeeting, #NLProc, #ACL2024 #arabicnlp #arabic
W/ @MaramHasanain, @_aridhasan @wzaghouani, @rasuwaileh
Arid Hasan retweeted
You still have time to participate in ArAIEval shared tasks with the ArabicNLP conference co-located with ACL In Bangkok! We propose two tasks: : Unimodal (Text) Propagandistic Technique Detection
and Multimodal Propagandistic Memes Classification araieval.gitlab.io
Arid Hasan retweeted
Join us for the ArAIEval Shared Task at the ArabicNLP Conference 2024, co-located with ACL 2024.
More: araieval.gitlab.io/
Data/Code: gitlab.com/araieval/araieval…
Registration:
Task 1: codalab.lisn.upsaclay.fr/com…
Task 2: codalab.lisn.upsaclay.fr/com…
#ArabicNLP #NLP #ACL2024 #ArAIEval #LanguageTechnology #Innovation #CallForParticipation #NLProc
@aclmeeting, #NLProc, #ACL2024 #arabicnlp #arabic
W/ @MaramHasanain, @_aridhasan @wzaghouani, @rasuwaileh
Arid Hasan retweeted
Join Arabic AI Tasks Evaluation (ArAIEval) for two tasks! 🌟 Task 1: Detect propagandistic techniques in text📝 Task 2: Classify propagandistic vs. non-propagandistic Arabic memes. 🖼️ Make an impact in Arabic NLP at ACL2024 in Bangkok. araieval.gitlab.io #NLProc @CHSS_HBKU

















