---
title: "Overview of the Shared Task on Multilevel Political Meme"
description: "This overview describes a DravidianLangTech–ACL 2026 shared task on automatically analysing… 18 references, cited by 6 papers here."
authors: ["Saranya Rajiakodi", "Shunmuga Priya Muthusamy Chinnan", "Premjith B", "Subalalitha CN", "Rahul Ponnusamy", "Anshid K A", "Bhuvaneswari Sivagnanam", "Bharathi Raja Chakravarthi"]
year: 2026
venue: "Proceedings of the Sixth Workshop on Speech, Vision, and Language Technologies for Dravidian Languages"
source: "https://aclanthology.org/2026.dravidianlangtech-1.12/"
doi: "10.18653/v1/2026.dravidianlangtech-1.12"
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/rajiakodi-2026-overview-shared-task-multilevel-political/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/rajiakodi-2026-overview-shared-task-multilevel-political/)

[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/3/) / Overview of the Shared Task on Multilevel Political Meme Classification in Tamil and Malayalam

# Overview of the Shared Task on Multilevel Political Meme Classification in Tamil and Malayalam

**Authors:** [Saranya Rajiakodi](https://md.research.elamur.ai/free/en/authors/saranya-rajiakodi/), [Shunmuga Priya Muthusamy Chinnan](https://md.research.elamur.ai/free/en/authors/shunmuga-priya-muthusamy-chinnan/), [Premjith B](https://md.research.elamur.ai/free/en/authors/premjith-b/), Subalalitha CN, [Rahul Ponnusamy](https://md.research.elamur.ai/free/en/authors/rahul-ponnusamy/), Anshid K A, [Bhuvaneswari Sivagnanam](https://md.research.elamur.ai/free/en/authors/bhuvaneswari-sivagnanam/) and [Bharathi Raja Chakravarthi](https://md.research.elamur.ai/free/en/authors/bharathi-raja-chakravarthi/) · **Year:** 2026 · **Venue:** [Proceedings of the Sixth Workshop on Speech, Vision, and Language Technologies for Dravidian Languages](https://md.research.elamur.ai/free/en/venues/dravidianlangtech/) ([Workshop on Speech, Vision, and Language Technologies for Dravidian Languages 2026](https://md.research.elamur.ai/free/en/venues/dravidianlangtech/2026/)) · **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/2026.dravidianlangtech-1.12/) · [PDF](https://aclanthology.org/2026.dravidianlangtech-1.12.pdf) · [DOI 10.18653/v1/2026.dravidianlangtech-1.12](https://doi.org/10.18653/v1/2026.dravidianlangtech-1.12) · [OpenAlex](https://openalex.org/W7166843268)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · [6](https://md.research.elamur.ai/free/en/papers/rajiakodi-2026-overview-shared-task-multilevel-political/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
This overview describes a DravidianLangTech–ACL 2026 shared task on automatically analysing political memes in Tamil and Malayalam. Both are low-resource languages where meaning depends on image, caption, sarcasm and local political culture.

The organisers defined a two-stage hierarchy:
- **Level 1:** decide whether a meme supports/praises or trolls/opposes a political entity.
- **Level 2:** given that stance, determine whether the target is an individual politician or a party/group.

Memes were scraped from Facebook, Instagram and Google using BeautifulSoup4 and platform APIs. They were then deduplicated, anonymised and annotated by native-speaker teams, with agreement measured via Krippendorff's α and Fleiss' κ. Submissions on Codabench were scored by macro-F1.

Of 64 registered teams, 19 submitted systems. Approaches included:
- multilingual text encoders combined with vision backbones;
- cross-modal fusion and cascaded architectures;
- OCR pipelines and fine-tuned vision-language models.

Stance detection proved fairly tractable, whereas target identification stayed difficult, particularly for Malayalam. The released benchmark offers a first systematic testbed for hierarchical multimodal political meme analysis in Dravidian languages.

## Key points
- **Dataset:** Tamil has 1003 memes (802 train / 201 test); Malayalam has 600 (500 train / 100 test). Troll memes outnumber supportive ones.
- **Inter-annotator agreement:**
  - Tamil: α = κ = 0.876 at Level 1 and 0.606 at Level 2.
  - Malayalam: α = 0.817 at Level 1 and 1.000 at Level 2.
- **Best average macro-F1:** 0.8029 for Tamil (IndiLangTech) and 0.7930 for Malayalam.
- **Level 1 vs. Level 2 gap:** Level 1 F1 reached about 0.96 for Malayalam and 0.93 for Tamil. The best Level 2 F1 was only about 0.62 for Malayalam and 0.69 for Tamil.
- **Models used:** text encoders included XLM-R, IndicBERT, MuRIL and mBERT; vision encoders included CLIP, ResNet and ConvNeXt. Some teams built image-only systems based on face detection and political symbol matching.

*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

- [Ollscoil na Gaillimhe – University of Galway](https://md.research.elamur.ai/free/en/universities/ollscoil-na-gaillimhe-university-of-galway/) — University, Galway, Ireland · 23 works · 2,229 citations
- [Amrita Vishwa Vidyapeetham](https://md.research.elamur.ai/free/en/universities/amrita-vishwa-vidyapeetham/) — University, Coimbatore, India · 7 works · 31 citations
- [Central University of Tamil Nadu](https://md.research.elamur.ai/free/en/universities/central-university-of-tamil-nadu/) — University, Thiruvarur, India · 3 works · 15 citations

📚 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 · [💭 Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/) · 1,552 works · [💭 Social Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/social-psychology/) · 306 works

🧭 Topics (2): [📏 Evaluation & benchmarks](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/) · 1,220 works · [🖼️ Vision & language](https://md.research.elamur.ai/free/en/topics/multimodal/) · 657 works

## Cite

Cite as: Saranya Rajiakodi, Shunmuga Priya Muthusamy Chinnan, Premjith B, Subalalitha CN, Rahul Ponnusamy, Anshid K A, Bhuvaneswari Sivagnanam and Bharathi Raja Chakravarthi. 2026. Overview of the Shared Task on Multilevel Political Meme Classification in Tamil and Malayalam. Proceedings of the Sixth Workshop on Speech, Vision, and Language Technologies for Dravidian Languages.

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

```bibtex
@inproceedings{rajiakodi2026overview,
  title = {{Overview of the Shared Task on Multilevel Political Meme Classification in Tamil and Malayalam}},
  author = {Saranya Rajiakodi, Shunmuga Priya Muthusamy Chinnan, Premjith B, Subalalitha CN, Rahul Ponnusamy, Anshid K A, Bhuvaneswari Sivagnanam and Bharathi Raja Chakravarthi},
  booktitle = {Proceedings of the Sixth Workshop on Speech, Vision, and Language Technologies for Dravidian Languages},
  year = {2026},
  doi = {10.18653/v1/2026.dravidianlangtech-1.12},
  url = {https://doi.org/10.18653/v1/2026.dravidianlangtech-1.12}
}
```

- [← previous MultiWikiQA: A Reading Comprehension Benchmark in 300+ Languages](https://md.research.elamur.ai/free/en/papers/smart-2026-multiwikiqa-reading-comprehension-benchmark-300/)
- [next → PDTrim: Targeted Pruning for Prefill-Decode Disaggregation in Inference](https://md.research.elamur.ai/free/en/papers/zhang-2026-pdtrim-targeted-pruning-prefill-decode/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/3/)

## References

18 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Shahad Abir. 2026. CYBER-PUNK@DravidianLangTech 2026: Multimodal Political Meme Classification using CLIP and Logo Similarity. In Proceedings of the Sixth Workshop on Speech, Vision, and Language Technologies for Dravidian Languages. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Shahad+Abir%2E+2026%2E+CYBER%2DPUNK%40DravidianLangTech+2026%3A+Multimodal+Political+Meme+Classification+using+CLIP+and+Logo+Similarity%2E+In+Proceedings+of+the+Sixth+Workshop+on+Speech%2C+Vision%2C+and+Language+Tech)
2. Mohammad Awad AlAfnan. 2025. The role of memes in shaping political discourse on social media. Studies in Media and Communication, 13(2):1–10. [find](https://scholar.google.com/scholar?q=Mohammad+Awad+AlAfnan%2E+2025%2E+The+role+of+memes+in+shaping+political+discourse+on+social+media%2E+Studies+in+Media+and+Communication%2C+13%282%29%3A1%E2%80%9310%2E)
3. Glenn Anderau and Daniel Barbarrusa. 2024. The function of memes in political discourse. Topoi, 43(5):1529–1546. [find](https://scholar.google.com/scholar?q=Glenn+Anderau+and+Daniel+Barbarrusa%2E+2024%2E+The+function+of+memes+in+political+discourse%2E+Topoi%2C+43%285%29%3A1529%E2%80%931546%2E)
4. Bharathi Raja Chakravarthi, Ruba Priyadharshini, Vigneshwaran Muralidaran, Shardul Suryawanshi, Navya Jose, Elizabeth Sherly, and John P McCrae. 2020. Overview of the track on sentiment analysis for dravidian languages in code-mixed text. In Proceedings of the 12th Annual Meeting of the Forum for In… [find](https://scholar.google.com/scholar?q=Bharathi+Raja+Chakravarthi%2C+Ruba+Priyadharshini%2C+Vigneshwaran+Muralidaran%2C+Shardul+Suryawanshi%2C+Navya+Jose%2C+Elizabeth+Sherly%2C+and+John+P+McCrae%2E+2020%2E+Overview+of+the+track+on+sentiment+analysis+for+d)
5. Bharathi Raja Chakravarthi, Saranya Rajiakodi, Rahul Ponnusamy, Kathiravan Pannerselvam, Ramachandran Rajalakshmi, Hariharan LekshmiAmmal, Anshid Kizhakkeparambil, Susminu S Kumar, Bhuvaneswari Sivagnanam, Charmathi Rajkumar, and 1 others. 2024. Overview of shared task on multitask meme classificati… [find](https://scholar.google.com/scholar?q=Bharathi+Raja+Chakravarthi%2C+Saranya+Rajiakodi%2C+Rahul+Ponnusamy%2C+Kathiravan+Pannerselvam%2C+Ramachandran+Rajalakshmi%2C+Hariharan+LekshmiAmmal%2C+Anshid+Kizhakkeparambil%2C+Susminu+S+Kumar%2C+Bhuvaneswari+Sivagn)
6. Arka Dutta, Anindya Majumder, Adnan Faisal, and Hasan Murad. 2026. CUET-2567@DravidianLangTech-ACL 2026: Multimodal Stance and Target Identification in Dravidian Political Memes. In Proceedings of the Sixth Workshop on Speech, Vision, and Language Technologies for Dravidian Languages. Association fo… [find](https://scholar.google.com/scholar?q=Arka+Dutta%2C+Anindya+Majumder%2C+Adnan+Faisal%2C+and+Hasan+Murad%2E+2026%2E+CUET%2D2567%40DravidianLangTech%2DACL+2026%3A+Multimodal+Stance+and+Target+Identification+in+Dravidian+Political+Memes%2E+In+Proceedings+of+the)
7. Eftekhar Hossain, Omar Sharif, and Mohammed Moshiul Hoque. 2022. Memosen: A multimodal dataset for sentiment analysis of memes. In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 1542–1554. [find](https://scholar.google.com/scholar?q=Eftekhar+Hossain%2C+Omar+Sharif%2C+and+Mohammed+Moshiul+Hoque%2E+2022%2E+Memosen%3A+A+multimodal+dataset+for+sentiment+analysis+of+memes%2E+In+Proceedings+of+the+Thirteenth+Language+Resources+and+Evaluation+Confe)
8. Md. Ajwad Hossain. 2026. RMS@DravidianLangTech 2026: Multimodal Gated Fusion for Hierarchical Tamil Political Meme Classification. In Proceedings of the Sixth Workshop on Speech, Vision, and Language Technologies for Dravidian Languages. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Md%2E+Ajwad+Hossain%2E+2026%2E+RMS%40DravidianLangTech+2026%3A+Multimodal+Gated+Fusion+for+Hierarchical+Tamil+Political+Meme+Classification%2E+In+Proceedings+of+the+Sixth+Workshop+on+Speech%2C+Vision%2C+and+Language+)
9. Marvin John Ignacio, Thanh Tin Nguyen, Hulin Jin, and Yong-Guk Kim. 2024. Meme analysis using llm-based contextual information and u-net encapsulated transformer. IEEE Access, 12:125993–126005. [find](https://scholar.google.com/scholar?q=Marvin+John+Ignacio%2C+Thanh+Tin+Nguyen%2C+Hulin+Jin%2C+and+Yong%2DGuk+Kim%2E+2024%2E+Meme+analysis+using+llm%2Dbased+contextual+information+and+u%2Dnet+encapsulated+transformer%2E+IEEE+Access%2C+12%3A125993%E2%80%93126005%2E)
10. Sixth Workshop on Speech, Vision, and Language Technologies for Dravidian Languages. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Sixth+Workshop+on+Speech%2C+Vision%2C+and+Language+Technologies+for+Dravidian+Languages%2E+Association+for+Computational+Linguistics%2E)
11. Hariram Veeramani, Surendrabikram Thapa, Mohammad Salman, Mohammad Ali Moni, and Usman Naseem. 2024. Umml: Layout-aware text-image fusion for unified multilingual multimodal learning. In 2024 IEEE International Conference on Data Mining Workshops (ICDMW), pages 174–182. IEEE. [find](https://scholar.google.com/scholar?q=Hariram+Veeramani%2C+Surendrabikram+Thapa%2C+Mohammad+Salman%2C+Mohammad+Ali+Moni%2C+and+Usman+Naseem%2E+2024%2E+Umml%3A+Layout%2Daware+text%2Dimage+fusion+for+unified+multilingual+multimodal+learning%2E+In+2024+IEEE+Int)
12. Karnati Vivek, Sai Mani Garrepalli, PVS Ajay Krishna, Inti Charan Kumar, Ravi Teja Gedela, and Sasib-hushara Rao Pappu. 2025. Telmemood: A multimodal dataset for sentiment analysis of telugu memes. In 2025 3rd International Conference on Intelligent Systems, Advanced Computing and Communication (ISA… [find](https://scholar.google.com/scholar?q=Karnati+Vivek%2C+Sai+Mani+Garrepalli%2C+PVS+Ajay+Krishna%2C+Inti+Charan+Kumar%2C+Ravi+Teja+Gedela%2C+and+Sasib%2Dhushara+Rao+Pappu%2E+2025%2E+Telmemood%3A+A+multimodal+dataset+for+sentiment+analysis+of+telugu+memes%2E+In)
13. Suriya K P, Durai Singh K, Gnanasabesan G, Ganesh Sundhar, Hari Krishnan N, and Jyothish Lal G. 2026. SYNAPSE@DravidianLangTech 2026: Multi-Level Political Meme Classification for Tamil and Malayalam. In Proceedings of the Sixth Workshop on Speech, Vision, and Language Technologies for Dravidian Lan… [find](https://scholar.google.com/scholar?q=Suriya+K+P%2C+Durai+Singh+K%2C+Gnanasabesan+G%2C+Ganesh+Sundhar%2C+Hari+Krishnan+N%2C+and+Jyothish+Lal+G%2E+2026%2E+SYNAPSE%40DravidianLangTech+2026%3A+Multi%2DLevel+Political+Meme+Classification+for+Tamil+and+Malayalam%2E)
14. Saurabh Kumar, Vivekananda G, Sanasam Ranbir Singh, and Sukumar Nandi. 2026. Indi-LangTech@DravidianLangTech 2026: Hierarchical Modeling for Multi-Level Political Meme Classification. In Proceedings of the Sixth Workshop on Speech, Vision, and Language Technologies for Dravidian Languages. Associati… [find](https://scholar.google.com/scholar?q=Saurabh+Kumar%2C+Vivekananda+G%2C+Sanasam+Ranbir+Singh%2C+and+Sukumar+Nandi%2E+2026%2E+Indi%2DLangTech%40DravidianLangTech+2026%3A+Hierarchical+Modeling+for+Multi%2DLevel+Political+Meme+Classification%2E+In+Proceedings+o)
15. Rahul Ponnusamy, Saranya Rajiakodi, Bhuvaneswari Sivagnanam, Anshid Kizhakkeparambil, Dhruv Sharma, Paul Buitelaar, and Bharathi Raja Chakravarthi. 2026. Fuse-md: A culturally-aware multimodal model for detecting misogyny memes. Natural Language Processing Journal, page 100197. [find](https://scholar.google.com/scholar?q=Rahul+Ponnusamy%2C+Saranya+Rajiakodi%2C+Bhuvaneswari+Sivagnanam%2C+Anshid+Kizhakkeparambil%2C+Dhruv+Sharma%2C+Paul+Buitelaar%2C+and+Bharathi+Raja+Chakravarthi%2E+2026%2E+Fuse%2Dmd%3A+A+culturally%2Daware+multimodal+model+f)
16. B Premjith, Bharathi Raja Chakravarthi, Malliga Subramanian, B Bharathi, Soman Kp, V Dhanalakshmi, K Sreelakshmi, Arunaggiri Pandian, and Prasanna Kumaresan. 2022. Findings of the shared task on multimodal sentiment analysis and troll meme classification in dravidian languages. In Proceedings of the… [find](https://scholar.google.com/scholar?q=B+Premjith%2C+Bharathi+Raja+Chakravarthi%2C+Malliga+Subramanian%2C+B+Bharathi%2C+Soman+Kp%2C+V+Dhanalakshmi%2C+K+Sreelakshmi%2C+Arunaggiri+Pandian%2C+and+Prasanna+Kumaresan%2E+2022%2E+Findings+of+the+shared+task+on+multi)
17. Alicia Trelles-Villanueva, Rafael Carrasco-Polaino, and Miguel Ángel Sánchez de la Nieta. 2025. Humor, persuasion, and politics: The impact of memes on x during the 2024 u.s. presidential election campaign. VISUAL REVIEW. International Visual Culture Review / Revista Internacional de Cultura Visual,… [find](https://scholar.google.com/scholar?q=Alicia+Trelles%2DVillanueva%2C+Rafael+Carrasco%2DPolaino%2C+and+Miguel+%C3%81ngel+S%C3%A1nchez+de+la+Nieta%2E+2025%2E+Humor%2C+persuasion%2C+and+politics%3A+The+impact+of+memes+on+x+during+the+2024+u%2Es%2E+presidential+election+cam)
18. Junain Uddin, Rahul Datta, Taha Ibne Abdullah, and Hasan Murad. 2026. [Semantica@DravidianLangTech 2026: Vision-Language Models for Hierarchical Political Meme Classification in Tamil and Malayalam](https://md.research.elamur.ai/free/en/papers/uddin-2026-semantica-dravidianlangtech-2026-vision-language/). In Proceedings of the

*Updated 31 October 2026*