---
title: "TTLab at StanceEval-2026 — Verma 2026"
description: "CLASP-Ar is the TTLab entry to StanceEval-2026. The task is to decide whether an Arabic post from X supports a given target topic, opposes it, or takes…"
authors: ["Bhuvanesh Verma", "Ali Abusaleh", "Alexander Mehler"]
year: 2026
venue: "arXiv:2609.29733"
source: "https://arxiv.org/abs/2609.29733"
doi: "10.48550/arxiv.2609.29733"
arxiv: "2609.29733"
licence: "CC BY 4.0 (arXiv)"
licence_url: "http://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/verma-2026-ttlab-stanceeval-2026-cloze-style/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/verma-2026-ttlab-stanceeval-2026-cloze-style/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/10/) / TTLab at StanceEval-2026: A Cloze-Style Prompting Approach for Arabic-Language Stance Detection (CLASP-Ar)

# TTLab at StanceEval-2026: A Cloze-Style Prompting Approach for Arabic-Language Stance Detection (CLASP-Ar)

**Authors:** [Bhuvanesh Verma](https://md.research.elamur.ai/free/en/authors/bhuvanesh-verma/), [Ali Abusaleh](https://md.research.elamur.ai/free/en/authors/ali-abusaleh/), [Alexander Mehler](https://md.research.elamur.ai/free/en/authors/alexander-mehler/) · **Year:** 2026 · **Venue:** arXiv:2609.29733 · **Licence:** [CC BY 4.0 (arXiv)](http://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/2609.29733) · [PDF](https://arxiv.org/pdf/2609.29733) · [DOI 10.48550/arxiv.2609.29733](https://doi.org/10.48550/arxiv.2609.29733) · [arXiv 2609.29733](https://arxiv.org/abs/2609.29733) · [OpenAlex](https://openalex.org/W7214312746)

**Citations:** 0 citations (openalex-snapshot · 2026-10-14)

## Summary
CLASP-Ar is the TTLab entry to StanceEval-2026. The task is to decide whether an Arabic post from X supports a given target topic, opposes it, or takes no position. Earlier top systems for Arabic stance relied on multitask heads, ensembles or target-specific models. The authors instead train a single encoder, aubmindlab/bert-base-arabertv02-twitter, to fill a masked slot. The prompt contains the target, a predicted sentiment label and the text. The MLM head may only choose among three one-character label tokens.

Training has two stages:
- **Intermediate pre-training** on ArabicStance-X, with sentiment labels supplied by the SARF classifier.
- **Fine-tuning** on Mawqif-XT, extended with synthetic None examples and trained with a class-weighted loss.

The system scores F_avg2 of 71.36% on Track 1 (seen targets) and 74.14% on Track 2 (unseen targets). The authors credit the stronger unseen-target result mainly to the diverse pre-training topics. The work suggests that auxiliary signals can be added cheaply as prompt text rather than as extra task heads, keeping one transferable model.

## Key points
- **Prompt and labels:** the prompt is laid out as Target / Sentiment / Stance [MASK] / Text. A verbalizer maps Favor, Against and None to the tokens 'F', 'A' and 'N', and the softmax runs only over these three. This avoids class names being split into several subwords.
- **Data:**
  - Mawqif-XT has 3,502 training and 619 dev instances, and Favor heavily outnumbers None.
  - ArabicStance-X has 14,477 tweets over 17 topics, of which 11,500 were used for two epochs of intermediate pre-training.
  - 1,815 extra None examples were created by pairing texts with unrelated targets.
- **Sentiment as input:** adding sentiment raised dev F_avg2 by 1.38 points (84.90 → 86.28). Adding sarcasm as well gave the best F_avg3 (72.50), but sentiment alone was the most consistent gain.
- **Imbalance fixes:**
  - Class weighting lifted F_none by roughly 7 points and F_avg3 by roughly 2 points over the baseline, and by up to 9 F1 points on None per the conclusion.
  - The synthetic None examples mostly helped the Against class while hurting None.
- **Sentiment source:** SARF, a MARBERTv2-based multi-view model, beat a frozen XLM-RoBERTa sentiment model as the label provider (dev F_avg2 81.50 vs 80.96). The None class remains weak and unstable, with standard deviation above 10 points in some setups.

## Interesting ideas
- Restricting a masked-LM prediction to a tiny set of single-token labels turns classification into reuse of the pretrained output layer, so no new head is learned and the label decision stays readable as a token choice. — For Mova (learn): when an LLM or encoder teacher labels silver data, forcing its answer into a closed token set would make teacher verdicts comparable across witnesses and easy to audit.
- Feeding a predicted auxiliary label (sentiment) into the input as plain text, instead of training a parallel task head, lets one model use a side signal without architectural changes. — For Mova (frames): a cheap upstream polarity or irony guess could be passed as an explicit, logged feature into speech-act or implicature decisions, keeping the reason for each decision traceable.
- Building "no relation" examples by pairing an input with a mismatched target is an almost free negative generator. The paper also shows its cost: it sharpened one boundary but blurred the very class it was meant to help. — For Mova (learn): shuffled text–question or text–frame pairs could serve as negative controls in evaluation, while checking per-class effects rather than trusting the aggregate.
- An intermediate training pass on a broader, multi-topic dataset before task fine-tuning was credited with better transfer to unseen targets. — For Mova (answer): readers for fables or school tests could first be tuned on a varied pool of question types, so new topics are less of a shock.
- Reporting per-class F1 with multi-seed spread exposed that the minority class was highly volatile, which a single averaged score would hide. — For Mova (learn): evaluation reports for rare frame roles or rare UD relations should include seed variance, so apparent gains on small classes are not mistaken for real ones.

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

Fields: [Linguistics & languages](https://md.research.elamur.ai/free/en/shelves/linguistics/)

## Cite

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

```bibtex
@misc{verma2026ttlab,
  title = {{TTLab at StanceEval-2026: A Cloze-Style Prompting Approach for Arabic-Language Stance Detection (CLASP-Ar)}},
  author = {Verma, Bhuvanesh and Abusaleh, Ali and Mehler, Alexander},
  year = {2026},
  doi = {10.48550/arxiv.2609.29733},
  eprint = {2609.29733},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2609.29733}
}
```

- [← previous TTLab at Daleel 2026: STAR-Ar, Sequence Tagging for Argument Recognition in Arabic](https://md.research.elamur.ai/free/en/papers/verma-2026-ttlab-daleel-2026-star-ar/)
- [next → TTSVowelViz: A Tool for Visualising Text-to-Speech Model Training via Vowel Spaces](https://md.research.elamur.ai/free/en/papers/udawatta-2026-ttsvowelviz-tool-visualising-text-speech/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/10/)

## References

23 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Ali Abusaleh, Bhuvanesh Verma, and Alexander Mehler. 2026. TTLab at AraSentEval: okSARF sentiment analysis via root-based fusion for multi-dialectal Arabic. In The 7th Workshop on Open-Source Arabic Corpora and Processing Tools (OSACT7) with 5 Shared Tasks, pages 262–268, Palma, Mallorca (Spain). Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Ali+Abusaleh%2C+Bhuvanesh+Verma%2C+and+Alexander+Mehler%2E+2026%2E+TTLab+at+AraSentEval%3A+okSARF+sentiment+analysis+via+root%2Dbased+fusion+for+multi%2Ddialectal+Arabic%2E+In+The+7th+Workshop+on+Open%2DSource+Arabic+C)
2. Rasha Albalawi, Nuha Albadi, Hamzah Luqman, Saad Ezzini, Maram Kurdi, Asma Yamani, Ahmed Ashraf, Maged Al-Shaibani, and Nora Alturayeif. 2026a. Stanceeval-2026: The second stance detection shared task. In Proceedings of the Fourth Arabic Natural Language Processing Conference (ArabicNLP 2026), Budapest, Hungary. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Rasha+Albalawi%2C+Nuha+Albadi%2C+Hamzah+Luqman%2C+Saad+Ezzini%2C+Maram+Kurdi%2C+Asma+Yamani%2C+Ahmed+Ashraf%2C+Maged+Al%2DShaibani%2C+and+Nora+Alturayeif%2E+2026a%2E+Stanceeval%2D2026%3A+The+second+stance+detection+shared+task)
3. Rasha Albalawi, Nuha Albadi, Hamzah Luqman, Maram Kurdi, Saad Ezzini, Asma Yamani, and Ahmed Ashraf. 2026b. Mawqif-xt: An arabic benchmark dataset for cross-target stance detection. Preprint, arXiv:2608.09539. [↗](https://arxiv.org/abs/2608.09539)
4. Mamoun Alghaslan and Khaled Almutairy. 2024. MGKM at StanceEval2024 fine-tuning large language models for Arabic stance detection. In Proceedings of the Second Arabic Natural Language Processing Conference, pages 816–822, Bangkok, Thailand. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Mamoun+Alghaslan+and+Khaled+Almutairy%2E+2024%2E+MGKM+at+StanceEval2024+fine%2Dtuning+large+language+models+for+Arabic+stance+detection%2E+In+Proceedings+of+the+Second+Arabic+Natural+Language+Processing+Confe) [our card](https://md.research.elamur.ai/free/en/papers/alghaslan-2024-mgkm-stanceeval2024-fine-tuning-large/)
5. Ali Alkhathlan, Faris Alahmadi, Faris Kateb, and Hend Al-Khalifa. 2025. Constructing and evaluating ArabicStanceX: a social media dataset for arabic stance detection. Front. Artif. Intell., 8:1615800. [find](https://scholar.google.com/scholar?q=Ali+Alkhathlan%2C+Faris+Alahmadi%2C+Faris+Kateb%2C+and+Hend+Al%2DKhalifa%2E+2025%2E+Constructing+and+evaluating+ArabicStanceX%3A+a+social+media+dataset+for+arabic+stance+detection%2E+Front%2E+Artif%2E+Intell%2E%2C+8%3A1615800%2E)
6. Nora Alturayeif, Hamzah Luqman, Zaid Alyafeai, and Asma Yamani. 2024. Stanceeval 2024: The first arabic stance detection shared task. In Proceedings of the Second Arabic Natural Language Processing Conference, pages 774–782. [find](https://scholar.google.com/scholar?q=Nora+Alturayeif%2C+Hamzah+Luqman%2C+Zaid+Alyafeai%2C+and+Asma+Yamani%2E+2024%2E+Stanceeval+2024%3A+The+first+arabic+stance+detection+shared+task%2E+In+Proceedings+of+the+Second+Arabic+Natural+Language+Processing+Co) [our card](https://md.research.elamur.ai/free/en/papers/alturayeif-2024-stanceeval-2024-first-arabic-stance/)
7. Nora Saleh Alturayeif, Hamzah Abdullah Luqman, and Moataz Aly Kamaleldin Ahmed. 2022. Mawqif: A multi-label Arabic dataset for target-specific stance detection. In Proceedings of the The Seventh Arabic Natural Language Processing Workshop (WANLP), pages 174–184, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Nora+Saleh+Alturayeif%2C+Hamzah+Abdullah+Luqman%2C+and+Moataz+Aly+Kamaleldin+Ahmed%2E+2022%2E+Mawqif%3A+A+multi%2Dlabel+Arabic+dataset+for+target%2Dspecific+stance+detection%2E+In+Proceedings+of+the+The+Seventh+Arabi)
8. Wissam Antoun, Fady Baly, and Hazem Hajj. 2020. Arabert: Transformer-based model for arabic language understanding. In LREC 2020 Workshop Language Resources and Evaluation Conference 11–16 May 2020, page 9. [find](https://scholar.google.com/scholar?q=Wissam+Antoun%2C+Fady+Baly%2C+and+Hazem+Hajj%2E+2020%2E+Arabert%3A+Transformer%2Dbased+model+for+arabic+language+understanding%2E+In+LREC+2020+Workshop+Language+Resources+and+Evaluation+Conference+11%E2%80%9316+May+2020%2C+p) [our card](https://md.research.elamur.ai/free/en/papers/antoun-2020-arabert-transformer-based-model-arabic/)
9. Mohamed Badran, Mo’men Hamdy, Marwan Torki, and Nagwa M El-Makky. 2024. Alexunlp-bh at stanceeval2024: Multiple contrastive losses ensemble strategy with multi-task learning for stance detection in arabic. In Proceedings of the Second Arabic Natural Language Processing Conference, pages 823–827. [find](https://scholar.google.com/scholar?q=Mohamed+Badran%2C+Mo%E2%80%99men+Hamdy%2C+Marwan+Torki%2C+and+Nagwa+M+El%2DMakky%2E+2024%2E+Alexunlp%2Dbh+at+stanceeval2024%3A+Multiple+contrastive+losses+ensemble+strategy+with+multi%2Dtask+learning+for+stance+detection+in+ar)
10. Roy Bar-Haim, Indrajit Bhattacharya, Francesco Din-uzzo, Amrita Saha, and Noam Slonim. 2017. Stance classification of context-dependent claims. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, pages 251–261. [find](https://scholar.google.com/scholar?q=Roy+Bar%2DHaim%2C+Indrajit+Bhattacharya%2C+Francesco+Din%2Duzzo%2C+Amrita+Saha%2C+and+Noam+Slonim%2E+2017%2E+Stance+classification+of+context%2Ddependent+claims%2E+In+Proceedings+of+the+15th+Conference+of+the+European+Ch)
11. Francesco Barbieri, Luis Espinosa Anke, and Jose Camacho-Collados. 2022. XLM-T: Multilingual language models in Twitter for sentiment analysis and beyond. In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 258–266, Marseille, France. European Language Resources Association. [find](https://scholar.google.com/scholar?q=Francesco+Barbieri%2C+Luis+Espinosa+Anke%2C+and+Jose+Camacho%2DCollados%2E+2022%2E+XLM%2DT%3A+Multilingual+language+models+in+Twitter+for+sentiment+analysis+and+beyond%2E+In+Proceedings+of+the+Thirteenth+Language+Res) [our card](https://md.research.elamur.ai/free/en/papers/barbieri-2022-xlm-t-multilingual-language-models/)
12. Saptarshi Ghosh, Kripabandhu Ghosh, Debasis Ganguly, Tanmoy Chakraborty, Gareth JF Jones, Marie-Francine Moens, and Muhammad Imran. 2018. Exploitation of social media for emergency relief and preparedness: Recent research and trends. Information Systems Frontiers, 20(5):901–907. [find](https://scholar.google.com/scholar?q=Saptarshi+Ghosh%2C+Kripabandhu+Ghosh%2C+Debasis+Ganguly%2C+Tanmoy+Chakraborty%2C+Gareth+JF+Jones%2C+Marie%2DFrancine+Moens%2C+and+Muhammad+Imran%2E+2018%2E+Exploitation+of+social+media+for+emergency+relief+and+prepared)
13. Ahmed Hasanaath and Aisha Alansari. 2024. Stancecrafters at stanceeval2024: Multi-task stance detection using bert ensemble with attention based aggregation. In Proceedings of the Second Arabic Natural Language Processing Conference, pages 811–815. [find](https://scholar.google.com/scholar?q=Ahmed+Hasanaath+and+Aisha+Alansari%2E+2024%2E+Stancecrafters+at+stanceeval2024%3A+Multi%2Dtask+stance+detection+using+bert+ensemble+with+attention+based+aggregation%2E+In+Proceedings+of+the+Second+Arabic+Natura) [our card](https://md.research.elamur.ai/free/en/papers/hasanaath-2024-stancecrafters-stanceeval2024-multi-task-stance/)
14. Emily Kubin and Christian Von Sikorski. 2021. The role of (social) media in political polarization: a systematic review. Annals of the International Communication Association, 45(3):188–206. [find](https://scholar.google.com/scholar?q=Emily+Kubin+and+Christian+Von+Sikorski%2E+2021%2E+The+role+of+%28social%29+media+in+political+polarization%3A+a+systematic+review%2E+Annals+of+the+International+Communication+Association%2C+45%283%29%3A188%E2%80%93206%2E)
15. Veronica Lynn, Salvatore Giorgi, Niranjan Balasubramanian, and H Andrew Schwartz. 2019. Tweet classification without the tweet: An empirical examination of user versus document attributes. In Proceedings of the third workshop on natural language processing and computational social science, pages 18–28. [find](https://scholar.google.com/scholar?q=Veronica+Lynn%2C+Salvatore+Giorgi%2C+Niranjan+Balasubramanian%2C+and+H+Andrew+Schwartz%2E+2019%2E+Tweet+classification+without+the+tweet%3A+An+empirical+examination+of+user+versus+document+attributes%2E+In+Proceedi)
16. Saif Mohammad, Svetlana Kiritchenko, Parinaz Sobhani, Xiaodan Zhu, and Colin Cherry. 2016. Semeval-2016 task 6: Detecting stance in tweets. In Proceedings of the 10th international workshop on semantic evaluation (SemEval-2016), pages 31–41. [find](https://scholar.google.com/scholar?q=Saif+Mohammad%2C+Svetlana+Kiritchenko%2C+Parinaz+Sobhani%2C+Xiaodan+Zhu%2C+and+Colin+Cherry%2E+2016%2E+Semeval%2D2016+task+6%3A+Detecting+stance+in+tweets%2E+In+Proceedings+of+the+10th+international+workshop+on+semanti) [our card](https://md.research.elamur.ai/free/en/papers/mohammad-2016-semeval-2016-task-6-detecting/)
17. Kamalich Muniz-Rodriguez, Sylvia K Ofori, Lauren C Bayliss, Jessica S Schwind, Kadiatou Diallo, Manyun Liu, Jingjing Yin, Gerardo Chowell, and Isaac Chun-Hai Fung. 2020. Social media use in emergency response to natural disasters: a systematic review with a public health perspective. Disaster medicine and public health preparedness, 14(1):139–149. [find](https://scholar.google.com/scholar?q=Kamalich+Muniz%2DRodriguez%2C+Sylvia+K+Ofori%2C+Lauren+C+Bayliss%2C+Jessica+S+Schwind%2C+Kadiatou+Diallo%2C+Manyun+Liu%2C+Jingjing+Yin%2C+Gerardo+Chowell%2C+and+Isaac+Chun%2DHai+Fung%2E+2020%2E+Social+media+use+in+emergency+)
18. Timo Schick and Hinrich Schütze. 2021. Exploiting cloze-questions for few-shot text classification and natural language inference. In Proceedings of the 16th conference of the European chapter of the association for computational linguistics: main volume, pages 255–269. [find](https://scholar.google.com/scholar?q=Timo+Schick+and+Hinrich+Sch%C3%BCtze%2E+2021%2E+Exploiting+cloze%2Dquestions+for+few%2Dshot+text+classification+and+natural+language+inference%2E+In+Proceedings+of+the+16th+conference+of+the+European+chapter+of+the+) [our card](https://md.research.elamur.ai/free/en/papers/schick-2021-exploiting-cloze-questions-few-shot/)
19. Jay J Van Bavel, Steve Rathje, Elizabeth Harris, Claire Robertson, and Anni Sternisko. 2021. How social media shapes polarization. Trends in cognitive sciences, 25(11):913–916. [find](https://scholar.google.com/scholar?q=Jay+J+Van+Bavel%2C+Steve+Rathje%2C+Elizabeth+Harris%2C+Claire+Robertson%2C+and+Anni+Sternisko%2E+2021%2E+How+social+media+shapes+polarization%2E+Trends+in+cognitive+sciences%2C+25%2811%29%3A913%E2%80%93916%2E)
20. Wan Wei, Xiao Zhang, Xuqin Liu, Wei Chen, and Tengjiao Wang. 2016. pkudblab at semeval-2016 task 6: A specific convolutional neural network system for effective stance detection. In Proceedings of the 10th international workshop on semantic evaluation (SemEval-2016), pages 384–388. [find](https://scholar.google.com/scholar?q=Wan+Wei%2C+Xiao+Zhang%2C+Xuqin+Liu%2C+Wei+Chen%2C+and+Tengjiao+Wang%2E+2016%2E+pkudblab+at+semeval%2D2016+task+6%3A+A+specific+convolutional+neural+network+system+for+effective+stance+detection%2E+In+Proceedings+of+the) [our card](https://md.research.elamur.ai/free/en/papers/wei-2016-pkudblab-semeval-2016-task-6/)
21. Brandon T Willard and Rémi Louf. 2023. Efficient guided generation for large language models. arXiv preprint arXiv:2307.09702. [↗](https://arxiv.org/abs/2307.09702)
22. Ruifeng Xu, Yu Zhou, Dongyin Wu, Lin Gui, Jiachen Du, and Yun Xue. 2016. Overview of nlpcc shared task 4: Stance detection in chinese microblogs. In International Conference on Computer Processing of Oriental Languages, pages 907–916. Springer. [find](https://scholar.google.com/scholar?q=Ruifeng+Xu%2C+Yu+Zhou%2C+Dongyin+Wu%2C+Lin+Gui%2C+Jiachen+Du%2C+and+Yun+Xue%2E+2016%2E+Overview+of+nlpcc+shared+task+4%3A+Stance+detection+in+chinese+microblogs%2E+In+International+Conference+on+Computer+Processing+of+)
23. Bowen Zhang, Genan Dai, Fuqiang Niu, Nan Yin, Xiaomao Fan, Senzhang Wang, Xiaochun Cao, and Hu Huang. 2024. A survey of stance detection on social media: New directions and perspectives. arXiv preprint arXiv:2409.15690. [↗](https://arxiv.org/abs/2409.15690)