---
title: "MultiMinds at MAHED 2025 — paper by Debnath et al. (2025)"
description: "The MultiMinds team competed in two subtasks of MAHED 2025, a shared task hosted… 75% accuracy. 31 references, cited by 1 paper here."
authors: ["Riddhiman Debnath", "Abdul Wadud Shakib", "Md Saiful Islam"]
year: 2025
venue: "Proceedings of The Third Arabic Natural Language Processing Conference: Shared Tasks"
source: "https://aclanthology.org/2025.arabicnlp-sharedtasks.93/"
doi: "10.18653/v1/2025.arabicnlp-sharedtasks.93"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/debnath-2025-multiminds-mahed-2025-multimodal-multitask/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/debnath-2025-multiminds-mahed-2025-multimodal-multitask/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2025](https://md.research.elamur.ai/free/en/papers/2025/page/15/) / MultiMinds at MAHED 2025: Multimodal and Multitask Approaches for Detecting Emotional, Hate, and Offensive Speech in Arabic Content

# MultiMinds at MAHED 2025: Multimodal and Multitask Approaches for Detecting Emotional, Hate, and Offensive Speech in Arabic Content

**Authors:** Riddhiman Debnath, Abdul Wadud Shakib and [Md Saiful Islam](https://md.research.elamur.ai/free/en/authors/md-saiful-islam/) · **Year:** 2025 · **Venue:** [Proceedings of The Third Arabic Natural Language Processing Conference: Shared Tasks](https://md.research.elamur.ai/free/en/venues/arabicnlp/) ([ArabicNLP 2025](https://md.research.elamur.ai/free/en/venues/arabicnlp/2025/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2025.arabicnlp-sharedtasks.93/) · [PDF](https://aclanthology.org/2025.arabicnlp-sharedtasks.93.pdf) · [DOI 10.18653/v1/2025.arabicnlp-sharedtasks.93](https://doi.org/10.18653/v1/2025.arabicnlp-sharedtasks.93) · [OpenAlex](https://openalex.org/W4416037749)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · [1](https://md.research.elamur.ai/free/en/papers/debnath-2025-multiminds-mahed-2025-multimodal-multitask/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
The MultiMinds team competed in two subtasks of MAHED 2025, a shared task hosted by ArabicNLP 2025 that addresses emotion, hate and offensive content in Arabic. Task 2 is multitask text classification: each post receives an emotion label (12 classes) plus offensive and hate flags. The team compared several systems on noisy, imbalanced data, which they first cleaned and orthographically normalized:
- an SVM baseline on TF-IDF features,
- XGBoost on TF-IDF or on fused AraBERT and XLM-RoBERTa [CLS] embeddings, tuned with Optuna,
- a fine-tuned AraBERTv2 with three task heads,
- gpt-oss-20b adapted with LoRA.

The multitask AraBERT gave the best average macro F1, about 0.68. Task 3 is hateful meme detection. Here a lightweight ELU-Net encoder–decoder processes the image, and cross-attention injects DistilBERT-tokenized caption features into its middle layers. With class weighting this reached roughly 75% accuracy. The team placed 10th in Task 2 and 7th in Task 3. The paper's main value is practical: it shows that class weighting and embedding fusion help, and that a small multimodal model is workable. It also shows plainly how severe imbalance can hide very weak minority-class performance.

## Key points
- **Task 2 data skew.** Offensive posts number 1744 vs 4216 non-offensive, and hateful posts 303 vs 1441. Emotions range from Anger (1551) down to Fear (53). The split was a stratified 70/15/15.
- **Task 2 average macro F1:**

  | System | Avg macro F1 |
  |---|---|
  | SVM + TF-IDF baseline | ≈0.51 |
  | Optuna-tuned, class-weighted XGBoost on fused AraBERT+XLM-RoBERTa | 0.574 |
  | Multitask AraBERTv2 | 0.684 (emotion only 0.267) |
  | DistilBERT | 0.683 |
  | gpt-oss-20b (post-competition) | 0.300 (emotion 0.014) |

- **Preprocessing.** The pipeline strips diacritics, repeated characters, Latin words and digits. It also maps the variant forms of Alif onto a single standard Alif (U+0627).
- **Task 3 data.** The meme set has 398 hateful vs 2663 non-hateful items. The baseline used CLIP ViT-B/32 features, first with an SVM and then with a 1D-CNN.
- **Best meme model.** ELU-Net + DistilBERT with class weights scored accuracy 0.754 and macro F1 0.500. However, F1 on the hateful class was only 0.165, against 0.858 on the non-hateful class. Training used about 6.2 GB of GPU memory on Colab, and early stopping halted at epoch 5.

## Interesting ideas
- One shared encoder feeding three small task-specific linear heads (emotion, offensive, hate) let related labels share a single representation, and it outperformed separate feature-based classifiers.
- Headline accuracy (75%) hid a hateful-class F1 of only 0.165, which shows that minority-class and macro scores must always be reported next to accuracy.
- Folding visually or phonetically interchangeable character variants into one canonical form before tokenization reduces spurious vocabulary splits in noisy social text.
- Hyperparameters were searched against a class-weighted objective that directly targets macro F1, rather than raw accuracy, which matters whenever labels are heavily skewed.
- Concatenating the [CLS] vectors of two different pretrained encoders gave a richer input than either encoder alone for a cheap downstream classifier.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (frames): one shared contextual encoder could feed separate light heads for speech act, irony and implicature cues, keeping each decision separately inspectable.
- **I2** (learn): evaluation reports should break scores down per rare category, such as rare UD relations or rare frame roles, so that majority-class wins cannot mask failures.
- **I3** (tokens): the en tokenizer could canonicalize curly vs straight apostrophes and quotes, dash variants and similar Unicode look-alikes before lemmatization, logging every mapping so it stays explainable.
- **I4** (learn): when training the perceptron or gparse on silver data, model selection could optimize macro scores over relation labels, keeping rare dependencies from being sacrificed.
- **I5** (tree): features from two different parser encoders could be fused for an auxiliary classifier, such as a dialect or label-confidence detector, without retraining either encoder.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Shahjalal University of Science and Technology](https://md.research.elamur.ai/free/en/universities/shahjalal-university-of-science-and-technology/) — University, Sylhet, Bangladesh · 2 works · 1 citation

Also on the byline: [Athabasca University](https://ror.org/01y3xgc52) · [Metropolitan University](https://ror.org/04hdrrs71).

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works

🧭 Topics (3): [😊 Sentiment & emotion](https://md.research.elamur.ai/free/en/topics/sentiment-emotion/) · 972 works · [Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/) · 2,908 works · [🖼️ Vision & language](https://md.research.elamur.ai/free/en/topics/multimodal/) · 657 works

## Cite

Cite as: Riddhiman Debnath, Abdul Wadud Shakib and Md Saiful Islam. 2025. MultiMinds at MAHED 2025: Multimodal and Multitask Approaches for Detecting Emotional, Hate, and Offensive Speech in Arabic Content. Proceedings of The Third Arabic Natural Language Processing Conference: Shared Tasks.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{debnath2025multiminds,
  title = {{MultiMinds at MAHED 2025: Multimodal and Multitask Approaches for Detecting Emotional, Hate, and Offensive Speech in Arabic Content}},
  author = {Riddhiman Debnath, Abdul Wadud Shakib and Md Saiful Islam},
  booktitle = {Proceedings of The Third Arabic Natural Language Processing Conference: Shared Tasks},
  year = {2025},
  doi = {10.18653/v1/2025.arabicnlp-sharedtasks.93},
  url = {https://doi.org/10.18653/v1/2025.arabicnlp-sharedtasks.93}
}
```

- [← previous Multi-Agent Interactive Question Generation Framework for Long Document Understanding](https://md.research.elamur.ai/free/en/papers/wang-2025-multi-agent-interactive-question-generation/)
- [next → N&N at QIAS 2025: Chain-of-Thought Ensembles with Retrieval-Augmented framework for Classical Arabic Islamic](https://md.research.elamur.ai/free/en/papers/alangari-2025-n-n-qias-2025-chain/)
- [↑ back to papers of 2025](https://md.research.elamur.ai/free/en/papers/2025/page/15/)

## References

31 references, 11 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Malak Abdullah, Mirsad Hadzikadicy, and Samira Shaikhz. 2018. Sedat: sentiment and emotion detection in arabic text using cnn-lstm deep learning. In 2018 17th IEEE international conference on machine learning and applications (ICMLA), pages 835–840. IEEE. [find](https://scholar.google.com/scholar?q=Malak+Abdullah%2C+Mirsad+Hadzikadicy%2C+and+Samira+Shaikhz%2E+2018%2E+Sedat%3A+sentiment+and+emotion+detection+in+arabic+text+using+cnn%2Dlstm+deep+learning%2E+In+2018+17th+IEEE+international+conference+on+machine+)
2. Lubna Al-Henaki, Hend Al-Khalifa, Abdulmalik Al-Salman, Hajar Alqubayshi, Hind Al-Twailay, Gheeda Alghamdi, and Hawra Aljasim. 2025. [Multiprose: A multi-label arabic dataset for propaganda, sentiment, and emotion detection](https://md.research.elamur.ai/free/en/papers/al-henaki-2025-multiprose-multi-label-arabic-dataset/). In International Conference on Applications of Natural Language to Information Systems, pages 156–172. Springer.
3. Firoj Alam, Md Rafiul Biswas, Uzair Shah, Wajdi Zaghouani, and Georgios Mikros. 2024a. [Propaganda to hate: A multimodal analysis of arabic memes with multi-agent llms](https://md.research.elamur.ai/free/en/papers/alam-2024-propaganda-hate-multimodal-analysis-arabic/). In International Conference on Web Information Systems Engineering, pages 380– 390. Springer. Firoj Alam, Abul Hasnat, Fatema Ahmad, Md.
4. Arid Hasan, and Maram Hasanain. 2024b. [ArMeme: Propagandistic content in Arabic memes](https://md.research.elamur.ai/free/en/papers/alam-2024-armeme-propagandistic-content-arabic-memes/). In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 21071–21090, Miami, Florida, USA. Association for Computational Linguistics.
5. Nouar AlDahoul and Yasir Zaki. 2025. [Detecting hope, hate, and emotion in arabic textual speech and multimodal memes using large language models](https://md.research.elamur.ai/free/en/papers/aldahoul-2025-detecting-hope-hate-emotion-arabic/). arXiv preprint arXiv:2508.15810. [↗](https://arxiv.org/abs/2508.15810)
6. Wassen Aldjanabi, Abdelghani Dahou, Mohammed AA Al-Qaness, Mohamed Abd Elaziz, Ahmed Mohamed Helmi, and Robertas Damaševičius. 2021. Arabic offensive and hate speech detection using a cross-corpora multi-task learning model. In Informatics, volume 8, page 69. MDPI. [↗](https://doi.org/10.3390/informatics8040069)
7. Safa Alsafari, Samira Sadaoui, and Malek Mouhoub. 2020. Hate and offensive speech detection on arabic social media. Online Social Networks and Media, 19:100096. [↗](https://doi.org/10.1016/j.osnem.2020.100096)
8. Kholoud Alsmearat, Mohammed Shehab, Mahmoud Al-Ayyoub, Riyad Al-Shalabi, and Ghassan Kanaan. 2015. Emotion analysis of arabic articles and its impact on identifying the author’s gender. In 2015 IEEE /ACS 12th International Conference of Computer Systems and Applications (AICCSA), pages 1–6. IEEE. [find](https://scholar.google.com/scholar?q=Kholoud+Alsmearat%2C+Mohammed+Shehab%2C+Mahmoud+Al%2DAyyoub%2C+Riyad+Al%2DShalabi%2C+and+Ghassan+Kanaan%2E+2015%2E+Emotion+analysis+of+arabic+articles+and+its+impact+on+identifying+the+author%E2%80%99s+gender%2E+In+2015+IEEE+%2F)
9. Nourah Alswaidan and Mohamed El Bachir Menai. 2020. Hybrid feature model for emotion recognition in arabic text. IEEE Access, 8:37843–37854. [↗](https://doi.org/10.1109/access.2020.2975906)
10. Wissam Antoun, Fady Baly, and Hazem Hajj. 2020. [Arabert: Transformer-based model for arabic language understanding](https://md.research.elamur.ai/free/en/papers/antoun-2020-arabert-transformer-based-model-arabic/). arXiv preprint arXiv:2003.00104. [↗](https://arxiv.org/abs/2003.00104)
11. Greeshma Arya, Mohammad Kamrul Hasan, Ashish Bagwari, Nurhizam Safie, Shayla Islam, Fatima Rayan Awad Ahmed, Aaishani De, Muhammad Attique Khan, and Taher M Ghazal. 2024. Multimodal hate speech detection in memes using contrastive language-image pre-training. IEEE Access, 12:22359–22375. Md. [↗](https://doi.org/10.1109/access.2024.3361322)
12. Rafiul Biswas and Wajdi Zaghouani. 2025a. [An annotated corpus of arabic tweets for hate speech analysis](https://md.research.elamur.ai/free/en/papers/zaghouani-2025-annotated-corpus-arabic-tweets-hate/). CoRR, abs/2505.11969. Md. [↗](https://arxiv.org/abs/2505.11969)
13. Rafiul Biswas and Wajdi Zaghouani. 2025b. [Emohopespeech: An annotated dataset of emotions and hope speech in english and arabic](https://md.research.elamur.ai/free/en/papers/zaghouani-2025-emohopespeech-annotated-dataset-emotions-hope/). CoRR, abs/2505.11959. [↗](https://arxiv.org/abs/2505.11959)
14. Hamdy Mubarak, Sabit Hassan, and Shammur Absar Chowdhury. 2023. [Emojis as anchors to detect arabic offensive language and hate speech](https://md.research.elamur.ai/free/en/papers/mubarak-2022-emojis-anchors-detect-arabic-offensive/). Natural Language Engineering, 29(6):1436–1457. [↗](https://doi.org/10.1017/s1351324923000402)
15. Anusha Chhabra and Dinesh Kumar Vishwakarma. 2023. A literature survey on multimodal and multilingual automatic hate speech identification. Multimedia Systems, 29(3):1203–1230. [↗](https://doi.org/10.1007/s00530-023-01051-8)
16. Andrew Perrin. 2015. Social media usage. Pew research center, 125:52–68. [find](https://scholar.google.com/scholar?q=Andrew+Perrin%2E+2015%2E+Social+media+usage%2E+Pew+research+center%2C+125%3A52%E2%80%9368%2E)
17. Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Unsupervised cross-lingual representation learning at scale. arXiv preprint arXiv:1911.02116. [↗](https://arxiv.org/abs/1911.02116)
18. Michael Han Daniel Han and Unsloth team. 2023. Unsloth. [find](https://scholar.google.com/scholar?q=Michael+Han+Daniel+Han+and+Unsloth+team%2E+2023%2E+Unsloth%2E)
19. Yunjiao Deng, Yulei Hou, Jiangtao Yan, and Daxing Zeng. 2022. Elu-net: An efficient and lightweight u-net for medical image segmentation. IEEE Access, 10:35932–35941. [↗](https://doi.org/10.1109/access.2022.3163711)
20. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2018. [BERT: pre-training of deep bidirectional transformers for language understanding](https://md.research.elamur.ai/free/en/papers/devlin-2019-bert-pre-training-deep-bidirectional/). CoRR, abs/1810.04805. [↗](https://arxiv.org/abs/1810.04805)
21. Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, and 1 others. 2021. [Learning transferable visual models from natural language supervision](https://md.research.elamur.ai/free/en/papers/radford-2021-learning-transferable-visual-models-natural/). In International conference on machine learning, pages 8748–8763. PMLR.
22. Victor Sanh, Lysandre Debut, Julien Chaumond, and Thomas Wolf. 2019. Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter. arXiv preprint arXiv:1910.01108. [↗](https://arxiv.org/abs/1910.01108)
23. Kuljeet Singh, Amit Mahajan, and Vibhakar Mansotra. 2021. 1d-cnn based model for classification and analysis of network attacks. International Journal of Advanced Computer Science and Applications, 12(11):604–613. [↗](https://doi.org/10.14569/ijacsa.2021.0121169)
24. Reza Drikvandi and Olamide Lawal. 2023. Sparse principal component analysis for natural language processing. Annals of data science, 10(1):25–41. [find](https://scholar.google.com/scholar?q=Reza+Drikvandi+and+Olamide+Lawal%2E+2023%2E+Sparse+principal+component+analysis+for+natural+language+processing%2E+Annals+of+data+science%2C+10%281%29%3A25%E2%80%9341%2E)
25. Wajdi Zaghouani, Md Rafiul Biswas, Mabrouka Bessghaier, Shimaa Ibrahim, Georgio Mikros, and Firoj Alam. 2025. Overview on mahed 2025 shared task: Multimodal detection of hope and hate emotions in arabic content. In The Third Arabic Natural Language Processing Conference (ArabicNLP 2025), Suzhou. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Wajdi+Zaghouani%2C+Md+Rafiul+Biswas%2C+Mabrouka+Bessghaier%2C+Shimaa+Ibrahim%2C+Georgio+Mikros%2C+and+Firoj+Alam%2E+2025%2E+Overview+on+mahed+2025+shared+task%3A+Multimodal+detection+of+hope+and+hate+emotions+in+arab)
26. Ahmed El-Sayed and Omar Nasr. 2024. Aast-nlp at multimodal hate speech event detection 2024: A multimodal approach for classification of text-embedded images based on clip and bert-based models. In Proceedings of the 7th Workshop on Challenges and Applications of Automated Extraction of Socio-political Events from Text (CASE 2024), pages 139–144. Wajdi Zaghouani, Hamdy Mubarak, and Md. [find](https://scholar.google.com/scholar?q=Ahmed+El%2DSayed+and+Omar+Nasr%2E+2024%2E+Aast%2Dnlp+at+multimodal+hate+speech+event+detection+2024%3A+A+multimodal+approach+for+classification+of+text%2Dembedded+images+based+on+clip+and+bert%2Dbased+models%2E+In+Pr)
27. Rafiul Biswas. 2024. [So hateful! building a multi-label hate speech annotated Arabic dataset](https://md.research.elamur.ai/free/en/papers/zaghouani-2024-so-hateful-building-multi-label/). In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 15044–15055, Torino, Italia. ELRA and ICCL.
28. Paula Fortuna and Sérgio Nunes. 2018. A survey on automatic detection of hate speech in text. Acm Computing Surveys (Csur), 51(4):1–30. [↗](https://doi.org/10.1145/3232676)
29. Amir Jalilifard, Vinicius Fernandes Caridá, Alex Fernandes Mansano, Rogers S Cristo, and Felipe Pen-horate Carvalho da Fonseca. 2021. Semantic sensitive tf-idf to determine word relevance in documents. In Advances in Computing and Network Communications: Proceedings of CoCoNet 2020, Volume 2, pages 327–337. Springer. [↗](https://doi.org/10.1007/978-981-33-6987-0_27)
30. Mohamed Bayan Kmainasi, Abul Hasnat, Md Arid Hasan, Ali Ezzat Shahroor, and Firoj Alam. 2025. Memeintel: Explainable detection of propagandistic and hateful memes. arXiv preprint arXiv:2502.16612. [↗](https://arxiv.org/abs/2502.16612)
31. Hongchan Li, Yantong Lu, and Haodong Zhu. 2024. Multi-modal sentiment analysis based on image and text fusion based on cross-attention mechanism. Electronics, 13(11):2069. [↗](https://doi.org/10.3390/electronics13112069)

*Updated 1 November 2026*