---
title: "blue at SMM4H-HeaRD 2026 — paper by Sharma et al. (2026)"
description: "Team blue entered six SMM4H-HeaRD 2026 shared tasks on health text from social… 17.0 points. 18 references, cited by 2 papers here."
authors: ["Krish Sharma", "Rhea Singhal", "Jatin Bedi"]
year: 2026
venue: "Proceedings of the 11th Social Media Mining for Health Research and Applications (SMM4H-HeaRD 2026) Workshop and Shared Tasks"
source: "https://aclanthology.org/2026.smm4h-1.13/"
doi: "10.18653/v1/2026.smm4h-1.13"
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/sharma-2026-blue-smm4h-heard-2026-class/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/sharma-2026-blue-smm4h-heard-2026-class/)

[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/7/) / blue at SMM4H-HeaRD 2026: Class-Weighted Transformer Ensembles with Structured Decoding and Chain-of-Thought Blending across Six Health NLP Shared Tasks

# blue at SMM4H-HeaRD 2026: Class-Weighted Transformer Ensembles with Structured Decoding and Chain-of-Thought Blending across Six Health NLP Shared Tasks

**Authors:** [Krish Sharma](https://md.research.elamur.ai/free/en/authors/krish-sharma/), [Rhea Singhal](https://md.research.elamur.ai/free/en/authors/rhea-singhal/) and [Jatin Bedi](https://md.research.elamur.ai/free/en/authors/jatin-bedi/) · **Year:** 2026 · **Venue:** [Proceedings of the 11th Social Media Mining for Health Research and Applications (SMM4H-HeaRD 2026) Workshop and Shared Tasks](https://md.research.elamur.ai/free/en/venues/smm4h/) ([Social Media Mining for Health Research and Applications Workshop and Shared Tasks 2026](https://md.research.elamur.ai/free/en/venues/smm4h/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.smm4h-1.13/) · [PDF](https://aclanthology.org/2026.smm4h-1.13.pdf) · [DOI 10.18653/v1/2026.smm4h-1.13](https://doi.org/10.18653/v1/2026.smm4h-1.13) · [OpenAlex](https://openalex.org/W7166861635)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · 2 in our library, of the 149,439 papers whose bibliographies we read

## Summary
Team blue entered six SMM4H-HeaRD 2026 shared tasks on health text from social media, PubMed, pathology reports and clinical documents. The tasks were adverse drug event detection in several languages, flu vaccine effectiveness from tweets, SARS-CoV-2 patient metadata, TNM cancer staging, opioid impact spans in Reddit posts, and multilingual clinical NER with annotation projection. They cover binary, multi-class, multi-label and BIO sequence-labelling setups. Instead of building six unrelated systems, the authors used one toolkit and adjusted it per task. It has three parts: inverse-frequency class weighting against strong label imbalance, ensembles over seeds or backbones to reduce variance, and post-hoc calibration of decision thresholds. Task-specific additions include:
- Viterbi-constrained BIO decoding,
- regex overrides for explicit pTNM mentions,
- blending with chain-of-thought predictions from Qwen2.5-7B-Instruct.

The best results were near-perfect TNM staging and flu vaccination classification above the organisers' 70B CoT baseline. The paper argues that careful fine-tuning of encoders can match or beat LLM prompting. It also warns that thresholds tuned on small dev sets do not transfer reliably to test data.

## Key points
- **Task 6 (TNM staging):** Clinical-BigBird with a 1536-token context, three classification heads and a regex override reached micro-F1 0.990 and macro-F1 0.938. The M-stage was the weakest component (F1 0.828).
- **Task 3 (flu vaccine effectiveness):** a 9-model ensemble (DeBERTa-v3-large, BERTweet-large and RoBERTa-large, each with 3 seeds) used posting dates prepended to the input and Qwen CoT outputs blended at weight 0.2.
  - Vaccination: µF1 0.872, above the baseline of 0.849.
  - Test result: µF1 0.918, below the baseline of 0.950.
- **Task 5 (patient metadata):** PubMedBERT with 5-fold CV and an out-of-fold threshold of 0.68 scored F1 0.764. This is far above Llama-3-70B prompting (0.558) and close to the BiomedBERT-Large benchmark (0.776).
- **Task 1 (ADE detection):** XLM-RoBERTa-large plus ruRoberta-large and cl-tohoku BERT specialists reached pooled F1 0.580 (baseline 0.525). Per-language thresholds overfit, with Russian dropping 17.0 points and Mandarin 13.1 points from dev to test.
- **Task 7 (opioid impact spans):** a ten-model Viterbi-decoded ensemble achieved relaxed F1 0.59 and strict F1 0.50, above the participant mean.
- **Task 8 (multilingual clinical NER):** XLM-RoBERTa-base trained jointly on 7 languages reached strict F1 of roughly 0.2–0.41.
- **Cross-task finding:** class weighting gave the largest single gains, for example +5.5 F1 on ADE detection and +5.2 AUROC on the M-stage.

*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

- [Thapar Institute of Engineering & Technology](https://md.research.elamur.ai/free/en/universities/thapar-institute-of-engineering-technology/) — University, Patiāla, India · 6 works · 11 citations

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [🧬 Biochemistry, Genetics and Molecular Biology](https://md.research.elamur.ai/free/en/shelves/biochemistry-genetics-and-molecular-biology/) · 263 works · [🧬 Molecular Biology](https://md.research.elamur.ai/free/en/shelves/biochemistry-genetics-and-molecular-biology/molecular-biology/) · 260 works

🧭 Topics (1): [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,685 works

## Cite

Cite as: Krish Sharma, Rhea Singhal and Jatin Bedi. 2026. blue at SMM4H-HeaRD 2026: Class-Weighted Transformer Ensembles with Structured Decoding and Chain-of-Thought Blending across Six Health NLP Shared Tasks. Proceedings of the 11th Social Media Mining for Health Research and Applications (SMM4H-HeaRD 2026) Workshop and Shared Tasks.

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

```bibtex
@inproceedings{sharma2026blue,
  title = {{blue at SMM4H-HeaRD 2026: Class-Weighted Transformer Ensembles with Structured Decoding and Chain-of-Thought Blending across Six Health NLP Shared Tasks}},
  author = {Krish Sharma, Rhea Singhal and Jatin Bedi},
  booktitle = {Proceedings of the 11th Social Media Mining for Health Research and Applications (SMM4H-HeaRD 2026) Workshop and Shared Tasks},
  year = {2026},
  doi = {10.18653/v1/2026.smm4h-1.13},
  url = {https://doi.org/10.18653/v1/2026.smm4h-1.13}
}
```

- [← previous BLEnD-Vis: Benchmarking Multimodal Cultural Understanding in Vision Language Models](https://md.research.elamur.ai/free/en/papers/tan-2026-blend-vis-benchmarking-multimodal-cultural/)
- [next → Bridging Scientific Heritage: An Arabic — Russian Parallel Corpus and LLM Benchmark for Sustainable Knowledge Transfer](https://md.research.elamur.ai/free/en/papers/arabov-2026-bridging-scientific-heritage-arabic-russian/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/7/)

## References

18 references, 6 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Jun Suzuki, Kentaro Sakuma, and Kentaro Inui. 2020. Pre-training of deep bidirectional transformers for Japanese. In Proc. Annual Conference of JSAI. [find](https://scholar.google.com/scholar?q=Jun+Suzuki%2C+Kentaro+Sakuma%2C+and+Kentaro+Inui%2E+2020%2E+Pre%2Dtraining+of+deep+bidirectional+transformers+for+Japanese%2E+In+Proc%2E+Annual+Conference+of+JSAI%2E)
2. Alexis Conneau, Kartikay Khandelwal, and 1 others. 2020. [Unsupervised cross-lingual representation learning at scale](https://md.research.elamur.ai/free/en/papers/conneau-2020-unsupervised-cross-lingual-representation-learning/). In Proc. ACL, pages 8440–8451.
3. Dongfang Xu, Guillermo Lopez García, Karen O’Connor, and 1 others. 2025. Mining social media data for influenza vaccine effectiveness using a large language model and chain-of-thought prompting. medRxiv. [find](https://scholar.google.com/scholar?q=Dongfang+Xu%2C+Guillermo+Lopez+Garc%C3%ADa%2C+Karen+O%E2%80%99Connor%2C+and+1+others%2E+2025%2E+Mining+social+media+data+for+influenza+vaccine+effectiveness+using+a+large+language+model+and+chain%2Dof%2Dthought+prompting%2E+medRx)
4. Sumon Kanti Dey, Jeanne M Powell, and 1 others. 2025. Inference gap in domain expertise and machine intelligence in named entity recognition. In Biocomputing 2026: Proc. Pacific Symposium, pages 12–26. [find](https://scholar.google.com/scholar?q=Sumon+Kanti+Dey%2C+Jeanne+M+Powell%2C+and+1+others%2E+2025%2E+Inference+gap+in+domain+expertise+and+machine+intelligence+in+named+entity+recognition%2E+In+Biocomputing+2026%3A+Proc%2E+Pacific+Symposium%2C+pages+12%E2%80%9326)
5. Manzil Zaheer, Guru Guruganesh, and 1 others. 2020. [Big bird: Transformers for longer sequences](https://md.research.elamur.ai/free/en/papers/zaheer-2020-big-bird-transformers-longer-sequences/). In NeurIPS, volume 33, pages 17283–17297.
6. Yu Gu, Robert Tinn, Hao Cheng, and 1 others. 2022. Domain-specific language model pretraining for biomedical NLP. ACM Trans. Computing for Healthcare, 3(1):1–23. [find](https://scholar.google.com/scholar?q=Yu+Gu%2C+Robert+Tinn%2C+Hao+Cheng%2C+and+1+others%2E+2022%2E+Domain%2Dspecific+language+model+pretraining+for+biomedical+NLP%2E+ACM+Trans%2E+Computing+for+Healthcare%2C+3%281%29%3A1%E2%80%9323%2E)
7. Daniil Zmitrovich, Alexander Abramov, and 1 others. 2023. RuRoBERTa: A pre-trained language model for Russian. arXiv:2309.10931. — [arXiv:2309.10931](https://arxiv.org/abs/2309.10931) [find](https://scholar.google.com/scholar?q=Daniil+Zmitrovich%2C+Alexander+Abramov%2C+and+1+others%2E+2023%2E+RuRoBERTa%3A+A+pre%2Dtrained+language+model+for+Russian%2E+arXiv%3A2309%2E10931%2E+%E2%80%94+%5BarXiv%3A2309%2E10931%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2309%2E10931%29)
8. Pengcheng He, Jianfeng Gao, and Weizhu Chen. 2023. [DeBERTaV3: Improving DeBERTa using ELECTRA-style pre-training with gradient-disentangled embedding sharing](https://md.research.elamur.ai/free/en/papers/he-2021-debertav3-improving-deberta-using-electra/). In Proc. ICLR.
9. Sarvnaz Karimi, Alejandro Metke-Jimenez, Madonna Kemp, and Chen Wang. 2015. CADEC: A corpus of adverse drug event annotations. J. Biomedical Informatics, 55:73–81. [find](https://scholar.google.com/scholar?q=Sarvnaz+Karimi%2C+Alejandro+Metke%2DJimenez%2C+Madonna+Kemp%2C+and+Chen+Wang%2E+2015%2E+CADEC%3A+A+corpus+of+adverse+drug+event+annotations%2E+J%2E+Biomedical+Informatics%2C+55%3A73%E2%80%9381%2E)
10. Jenna Kefeli, Jacob Berkowitz, and 1 others. 2024. Generalizable and automated classification of TNM stage from pathology reports with external validation. Nature Communications, 15(1):8916. [find](https://scholar.google.com/scholar?q=Jenna+Kefeli%2C+Jacob+Berkowitz%2C+and+1+others%2E+2024%2E+Generalizable+and+automated+classification+of+TNM+stage+from+pathology+reports+with+external+validation%2E+Nature+Communications%2C+15%281%29%3A8916%2E)
11. Woosuk Kwon, Zhuohan Li, and 1 others. 2023. [Efficient memory management for large language model serving with PagedAttention](https://md.research.elamur.ai/free/en/papers/kwon-2023-efficient-memory-management-large-language/). In Proc. SOSP, pages 611–626.
12. Yikuan Li, Ramsey M Wehbe, and 1 others. 2023. A comparative study of pretrained language models for long clinical text. JAMIA, 30(2):340–347. [find](https://scholar.google.com/scholar?q=Yikuan+Li%2C+Ramsey+M+Wehbe%2C+and+1+others%2E+2023%2E+A+comparative+study+of+pretrained+language+models+for+long+clinical+text%2E+JAMIA%2C+30%282%29%3A340%E2%80%93347%2E)
13. Yinhan Liu, Myle Ott, Naman Goyal, and 1 others. 2019. [RoBERTa: A robustly optimized BERT pretraining approach](https://md.research.elamur.ai/free/en/papers/liu-2019-roberta-robustly-optimized-bert-pretraining/). arXiv:1907.11692.
14. Ilya Loshchilov and Frank Hutter. 2019. Decoupled weight decay regularization. In Proc. ICLR. [find](https://scholar.google.com/scholar?q=Ilya+Loshchilov+and+Frank+Hutter%2E+2019%2E+Decoupled+weight+decay+regularization%2E+In+Proc%2E+ICLR%2E)
15. Rafael Müller, Simon Kornblith, and Geoffrey E Hinton. 2019. When does label smoothing help? In NeurIPS, volume 32. [find](https://scholar.google.com/scholar?q=Rafael+M%C3%BCller%2C+Simon+Kornblith%2C+and+Geoffrey+E+Hinton%2E+2019%2E+When+does+label+smoothing+help%3F+In+NeurIPS%2C+volume+32%2E)
16. Dat Quoc Nguyen, Thanh Vu, and Anh Tuan Nguyen. 2020. [BERTweet: A pre-trained language model for English tweets](https://md.research.elamur.ai/free/en/papers/nguyen-2020-bertweet-pre-trained-language-model/). In Proc. EMNLP: System Demonstrations, pages 9–14.
17. NLLB Team, Marta R Costa-jussà, and 1 others. 2022. No language left behind: Scaling human-centered machine translation. arXiv:2207.04672. [find](https://scholar.google.com/scholar?q=NLLB+Team%2C+Marta+R+Costa%2Djuss%C3%A0%2C+and+1+others%2E+2022%2E+No+language+left+behind%3A+Scaling+human%2Dcentered+machine+translation%2E+arXiv%3A2207%2E04672%2E)
18. Abeed Sarker and Graciela Gonzalez. 2015. Portable automatic text classification for adverse drug reaction detection via multi-corpus training. J. Biomedical Informatics, 53:196–207. [find](https://scholar.google.com/scholar?q=Abeed+Sarker+and+Graciela+Gonzalez%2E+2015%2E+Portable+automatic+text+classification+for+adverse+drug+reaction+detection+via+multi%2Dcorpus+training%2E+J%2E+Biomedical+Informatics%2C+53%3A196%E2%80%93207%2E)

*Updated 2 November 2026*