---
title: "YNU-HPCC at SemEval-2026 Task 6 — paper by Wen et al. (2026)"
description: "The paper describes the YNU-HPCC entry to SemEval-2026 Task 6, which asks systems to judge how clearly… 0.824 accuracy. 19 references."
authors: ["Canning Wen", "Jin Wang", "Xuejie Zhang"]
year: 2026
venue: "Proceedings of the 20th International Workshop on Semantic Evaluation (2026)"
source: "https://aclanthology.org/2026.semeval-1.45/"
doi: "10.18653/v1/2026.semeval-1.45"
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/wen-2026-ynu-hpcc-semeval-2026-task/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/wen-2026-ynu-hpcc-semeval-2026-task/)

[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/57/) / YNU-HPCC at SemEval-2026 Task 6: Hierarchical Taxonomy Prompting and CoT Distillation for Political Clarity Classification

# YNU-HPCC at SemEval-2026 Task 6: Hierarchical Taxonomy Prompting and CoT Distillation for Political Clarity Classification

**Authors:** Canning Wen, [Jin Wang](https://md.research.elamur.ai/free/en/authors/jin-wang/) and [Xuejie Zhang](https://md.research.elamur.ai/free/en/authors/xuejie-zhang/) · **Year:** 2026 · **Venue:** [Proceedings of the 20th International Workshop on Semantic Evaluation (2026)](https://md.research.elamur.ai/free/en/venues/semeval/) ([SemEval 2026](https://md.research.elamur.ai/free/en/venues/semeval/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.semeval-1.45/) · [PDF](https://aclanthology.org/2026.semeval-1.45.pdf) · [DOI 10.18653/v1/2026.semeval-1.45](https://doi.org/10.18653/v1/2026.semeval-1.45) · [OpenAlex](https://openalex.org/W7166814888)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
The paper describes the YNU-HPCC entry to SemEval-2026 Task 6, which asks systems to judge how clearly politicians answer interview questions. Subtask 1 uses three clarity labels: Clear Reply, Ambivalent, and Clear Non-Reply. Subtask 2 uses nine finer evasion-strategy labels. The authors argue that models trained or prompted to emit only a label capture shallow semantics, so they make the model reason before deciding.

They propose two approaches:
- **Hierarchical Taxonomy Prompting (HTP):** The prompt walks an LLM through the organizers' taxonomy as a two-stage decision tree, choosing the clarity level first and then the evasion type. It includes explicit tests for the ambivalent cases and requires written analysis before the label.
- **CoT distillation:** DeepSeek V3.1 generates reasoning chains on the QEvasion training set, and smaller Qwen3 and Llama 3.1 models are fine-tuned on them.

HTP with MiMo-V2-Flash achieved the best macro F1 on both subtasks, and distillation clearly helped the small models. The team placed 6th and 5th, suggesting that explicit reasoning matters for detecting rhetorical evasion.

## Key points
- **Data:** Only the organizers' QEvasion dataset (English) was used, with 3,450 training and 308 test examples. Evaluation is by macro F1, with accuracy reported as a secondary metric.
- **Best results (HTP with MiMo-V2-Flash):** macro F1 of 0.776 on Task 1 and 0.559 on Task 2. HTP beat zero-shot, one-shot, few-shot and plain CoT prompting.
- **Distilled models:** Llama-3.1-8B reached 0.824 accuracy and 0.753 F1 on Task 1. Among encoders, ModernBERT-large was strongest (0.633 F1 on Task 1).
- **Fine-tuning setup:** Student models were trained with trl and unsloth, using 4-bit quantization, LoRA with rank 16, a learning rate of 2e-4, and 3 epochs. The best sampling temperature for Llama-3.1-8B was 0.8.
- **Error analysis:** The distilled model often falls back on Deflection, misclassifying many Explicit, Implicit and Dodging replies as Deflection. Dodging and Implicit answers are the hardest classes to recognize.

*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

- [Yunnan University](https://md.research.elamur.ai/free/en/universities/yunnan-university/) — University, Kunming, China · 20 works · 40 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 · [🤝 Social Sciences](https://md.research.elamur.ai/free/en/shelves/social-sciences/) · 1,954 works · [🤝 General Social Sciences](https://md.research.elamur.ai/free/en/shelves/social-sciences/general-social-sciences/) · 254 works

🧭 Topics (2): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,894 works · [🪶 Efficient models & compression](https://md.research.elamur.ai/free/en/topics/efficient-models/) · 297 works

## Cite

Cite as: Canning Wen, Jin Wang and Xuejie Zhang. 2026. YNU-HPCC at SemEval-2026 Task 6: Hierarchical Taxonomy Prompting and CoT Distillation for Political Clarity Classification. Proceedings of the 20th International Workshop on Semantic Evaluation (2026).

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

```bibtex
@inproceedings{wen2026ynu,
  title = {{YNU-HPCC at SemEval-2026 Task 6: Hierarchical Taxonomy Prompting and CoT Distillation for Political Clarity Classification}},
  author = {Canning Wen, Jin Wang and Xuejie Zhang},
  booktitle = {Proceedings of the 20th International Workshop on Semantic Evaluation (2026)},
  year = {2026},
  doi = {10.18653/v1/2026.semeval-1.45},
  url = {https://doi.org/10.18653/v1/2026.semeval-1.45}
}
```

- [← previous YNU-HPCC at SemEval-2026 Task 13: Robust Machine-Generated Code Detection under Distribution Shifts](https://md.research.elamur.ai/free/en/papers/xing-2026-ynu-hpcc-semeval-2026-task/)
- [next → You Can Have a Second Chance: Unbiased and Multi-bit Watermarking for Diffusion Language Models with Regret-based Remasking](https://md.research.elamur.ai/free/en/papers/yang-2026-you-can-have-second-chance/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/57/)

## References

19 references, 12 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Ron Banner, Yury Nahshan, and Daniel Soudry. 2019. Post training 4-bit quantization of convolutional networks for rapid-deployment. Advances in neural information processing systems, 32. [find](https://scholar.google.com/scholar?q=Ron+Banner%2C+Yury+Nahshan%2C+and+Daniel+Soudry%2E+2019%2E+Post+training+4%2Dbit+quantization+of+convolutional+networks+for+rapid%2Ddeployment%2E+Advances+in+neural+information+processing+systems%2C+32%2E)
2. Kaiyuan Chen, Jin Wang, and Xuejie Zhang. 2024. Mathematical reasoning via multi-step self questioning and answering for small language models. In CCF International Conference on Natural Language Processing and Chinese Computing, pages 81–93. Springer. [find](https://scholar.google.com/scholar?q=Kaiyuan+Chen%2C+Jin+Wang%2C+and+Xuejie+Zhang%2E+2024%2E+Mathematical+reasoning+via+multi%2Dstep+self+questioning+and+answering+for+small+language+models%2E+In+CCF+International+Conference+on+Natural+Language+Proc)
3. Kaiyuan Chen, Jin Wang, and Xuejie Zhang. 2025. Learning to reason via self-iterative process feedback for small language models. In Proceedings of the 31st International Conference on Computational Linguistics, pages 3027–3042. [find](https://scholar.google.com/scholar?q=Kaiyuan+Chen%2C+Jin+Wang%2C+and+Xuejie+Zhang%2E+2025%2E+Learning+to+reason+via+self%2Diterative+process+feedback+for+small+language+models%2E+In+Proceedings+of+the+31st+International+Conference+on+Computational+L)
4. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. [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/). In Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies,…
5. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, and 1 others. 2024. [The llama 3 herd of models](https://md.research.elamur.ai/free/en/papers/grattafiori-2024-llama-3-herd-models/). arXiv preprint arXiv:2407.21783.
6. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Peiyi Wang, Qihao Zhu, Runxin Xu, Ruoyu Zhang, Shirong Ma, Xiao Bi, and 1 others. 2025. [Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2025-deepseek-r1-incentivizing-reasoning-capability/). arXiv preprint arXiv:2501.12948.
7. Pengcheng He, Xiaodong Liu, Jianfeng Gao, and Weizhu Chen. 2020. [Deberta: Decoding-enhanced bert with disentangled attention](https://md.research.elamur.ai/free/en/papers/he-2020-deberta-decoding-enhanced-bert-disentangled/). arXiv preprint arXiv:2006.03654.
8. Geoffrey Hinton, Oriol Vinyals, and Jeff Dean. 2015. [Distilling the knowledge in a neural network](https://md.research.elamur.ai/free/en/papers/hinton-2015-distilling-knowledge-neural-network/). arXiv preprint arXiv:1503.02531.
9. Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Liang Wang, Weizhu Chen, and 1 others. 2022. [Lora: Low-rank adaptation of large language models](https://md.research.elamur.ai/free/en/papers/hu-2021-lora-low-rank-adaptation-large/). Iclr, 1(2):3.
10. Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, and 1 others. 2024. Deepseek-v3 technical report. arXiv preprint arXiv:2412.19437. [find](https://scholar.google.com/scholar?q=Aixin+Liu%2C+Bei+Feng%2C+Bing+Xue%2C+Bingxuan+Wang%2C+Bochao+Wu%2C+Chengda+Lu%2C+Chenggang+Zhao%2C+Chengqi+Deng%2C+Chenyu+Zhang%2C+Chong+Ruan%2C+and+1+others%2E+2024%2E+Deepseek%2Dv3+technical+report%2E+arXiv+preprint+arXiv%3A2412)
11. Tiesunlong Shen, Rui Mao, Jin Wang, Heming Sun, Jian Zhang, Xuejie Zhang, and Erik Cambria. 2026. Llmdoctor: Token-level flow-guided preference optimization for efficient test-time alignment of large language models. arXiv preprint arXiv:2601.10416. — [arXiv:2601.10416](https://arxiv.org/abs/2601.10416) [find](https://scholar.google.com/scholar?q=Tiesunlong+Shen%2C+Rui+Mao%2C+Jin+Wang%2C+Heming+Sun%2C+Jian+Zhang%2C+Xuejie+Zhang%2C+and+Erik+Cambria%2E+2026%2E+Llmdoctor%3A+Token%2Dlevel+flow%2Dguided+preference+optimization+for+efficient+test%2Dtime+alignment+of+large+)
12. Konstantinos Thomas, Giorgos Filandrianos, Maria Lymperaiou, Chrysoula Zerva, and Giorgos Stamou. 2024. ” i never said that”: A dataset, taxonomy and baselines on response clarity classification. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 5204–5233. [find](https://scholar.google.com/scholar?q=Konstantinos+Thomas%2C+Giorgos+Filandrianos%2C+Maria+Lymperaiou%2C+Chrysoula+Zerva%2C+and+Giorgos+Stamou%2E+2024%2E+%E2%80%9D+i+never+said+that%E2%80%9D%3A+A+dataset%2C+taxonomy+and+baselines+on+response+clarity+classification%2E+In+F)
13. Konstantinos Thomas, Giorgos Filandrianos, Maria Lymperaiou, Chrysoula Zerva, and Giorgos Stamou. 2026. [Semeval-2026 task 6: Clarity – unmasking political question evasions](https://md.research.elamur.ai/free/en/papers/thomas-2026-semeval-2026-task-6-clarity/). Preprint, arXiv:2603.14027.
14. Benjamin Warner, Antoine Chaffin, Benjamin Clavié, Orion Weller, Oskar Hallström, Said Taghadouini, Alexis Gallagher, Raja Biswas, Faisal Ladhak, Tom Aarsen, and 1 others. 2025. [Smarter, better, faster, longer: A modern bidirectional encoder for fast, memory efficient, and long context finetuning an…](https://md.research.elamur.ai/free/en/papers/warner-2024-smarter-better-faster-longer-modern/)
15. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, and 1 others. 2022. [Chain-of-thought prompting elicits reasoning in large language models](https://md.research.elamur.ai/free/en/papers/wei-2022-chain-thought-prompting-elicits-reasoning/). Advances in neural information processing systems, 35:24824– 24837.
16. Bangjun Xiao, Bingquan Xia, Bo Yang, Bofei Gao, Bowen Shen, Chen Zhang, Chenhong He, Chiheng Lou, Fuli Luo, Gang Wang, and 1 others. 2026. [Mimo-v2-flash technical report](https://md.research.elamur.ai/free/en/papers/team-2026-llm-mimo-v2-flash-technical/). arXiv preprint arXiv:2601.02780.
17. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, and 1 others. 2025. [Qwen3 technical report](https://md.research.elamur.ai/free/en/papers/yang-2025-qwen3-technical-report/). arXiv preprint arXiv:2505.09388.
18. Guangmin Zheng, Jin Wang, Xiaobing Zhou, and Xuejie Zhang. 2024. Enhancing semantics in multimodal chain of thought via soft negative sampling. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 6059– 60… [find](https://scholar.google.com/scholar?q=Guangmin+Zheng%2C+Jin+Wang%2C+Xiaobing+Zhou%2C+and+Xuejie+Zhang%2E+2024%2E+Enhancing+semantics+in+multimodal+chain+of+thought+via+soft+negative+sampling%2E+In+Proceedings+of+the+2024+Joint+International+Conferenc)
19. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. [Roberta: A robustly optimized bert pretraining approach](https://md.research.elamur.ai/free/en/papers/liu-2019-roberta-robustly-optimized-bert-pretraining/). arXiv preprint arXiv:1907.11692.

*Updated 2 November 2026*