---
title: "Vulnerability of LLMs to Vertically Aligned Text Manipulations"
description: "The paper asks whether decoder-based LLMs are as vulnerable as encoder models like BERT when… 42 references, cited by 5 papers here."
authors: ["Zhecheng Li", "Yiwei Wang", "Bryan Hooi", "Yujun Cai", "Zhen Xiong", "Nanyun Peng", "Kai-Wei Chang"]
year: 2025
venue: "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)"
source: "https://aclanthology.org/2025.acl-long.978/"
doi: "10.18653/v1/2025.acl-long.978"
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/li-2025-vulnerability-llms-vertically-aligned-text/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/li-2025-vulnerability-llms-vertically-aligned-text/)

[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/11/) / Vulnerability of LLMs to Vertically Aligned Text Manipulations

# Vulnerability of LLMs to Vertically Aligned Text Manipulations

**Authors:** [Zhecheng Li](https://md.research.elamur.ai/free/en/authors/zhecheng-li/), [Yiwei Wang](https://md.research.elamur.ai/free/en/authors/yiwei-wang/), [Bryan Hooi](https://md.research.elamur.ai/free/en/authors/bryan-hooi/), [Yujun Cai](https://md.research.elamur.ai/free/en/authors/yujun-cai/), [Zhen Xiong](https://md.research.elamur.ai/free/en/authors/zhen-xiong/), [Nanyun Peng](https://md.research.elamur.ai/free/en/authors/nanyun-peng/) and [Kai-Wei Chang](https://md.research.elamur.ai/free/en/authors/kai-wei-chang/) · **Year:** 2025 · **Venue:** [Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2025](https://md.research.elamur.ai/free/en/venues/acl/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.acl-long.978/) · [PDF](https://aclanthology.org/2025.acl-long.978.pdf) · [DOI 10.18653/v1/2025.acl-long.978](https://doi.org/10.18653/v1/2025.acl-long.978) · [OpenAlex](https://openalex.org/W4412944893)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · [5](https://md.research.elamur.ai/free/en/papers/li-2025-vulnerability-llms-vertically-aligned-text/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
The paper asks whether decoder-based LLMs are as vulnerable as encoder models like BERT when some words in a text are written vertically, one character per line. People can still read such text easily. The risk is that this formatting could help harmful content slip past automated filters. To test it, the authors use a prompted LLM to pick the most meaningful words in each input, which is cheaper than the earlier greedy word-importance search. A grid-based procedure then renders those words vertically while the rest of the sentence stays horizontal. They evaluate closed-source models (GPT-3.5, GPT-4, GPT-4o, GPT-4o-mini) and open-source models (LLaMA 3/3.1, Gemma 2) on five classification datasets. Accuracy drops sharply for every model, often by 20–45 points. Chain-of-Thought prompting does not fix the problem. Few-shot prompting with hand-written analyses of vertical examples does help. Tokenizer and attention analyses explain the failure, which matters for moderation, spam and toxicity detection systems.

## Key points
- **Datasets:** SST-2, CoLA, QNLI, Rotten Tomatoes and Jigsaw Toxicity, covering sentiment, acceptability, QA entailment and toxic comment detection.
- **Size of the drops:**
  - GPT-4o fell from 95 to 68 on SST-2 and from 87 to 47 on CoLA.
  - LLaMA 3.1-8B fell from 93 to 51 on SST-2.
  - LLaMA 3-70B fell from 91 to 46 on Rotten Tomatoes.
- **Tokenization cause:** the LLaMA 3.1-8B tokenizer encodes "vertical" as one token when written normally, but as 15 tokens when written vertically, because of the added spaces and line breaks.
- **Attention cause:** attention maps of LLaMA 3.1-8B show that key cue words ("miserable", "idiot") lose their strong link to the label tokens "negative" and "toxic" once they are split vertically.
- **Mitigation and limits:** few-shot prompts with manually crafted analysis help (e.g., GPT-4 three-shot recovers the hidden words), while CoT does not. Fine-tuning and text generation tasks are left for future work.

*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

- [University of Southern California](https://md.research.elamur.ai/free/en/universities/university-of-southern-california/) — University, Los Angeles, United States · 162 works · 35,836 citations
- [National University of Singapore](https://md.research.elamur.ai/free/en/universities/national-university-of-singapore/) — University, Singapore, Singapore · 180 works · 34,304 citations
- [University of California, Los Angeles](https://md.research.elamur.ai/free/en/universities/university-of-california-los-angeles/) — University, Los Angeles, United States · 87 works · 54,961 citations
- [University of California San Diego](https://md.research.elamur.ai/free/en/universities/university-of-california-san-diego/) — University, San Diego, United States · 73 works · 51,948 citations
- [The University of Queensland](https://md.research.elamur.ai/free/en/universities/the-university-of-queensland/) — University, Brisbane, Australia · 14 works · 4,603 citations
- [University of California, Merced](https://md.research.elamur.ai/free/en/universities/university-of-california-merced/) — University, Merced, United States · 11 works · 425 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 · [🎭 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 (1): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,320 works

## Cite

Cite as: Zhecheng Li, Yiwei Wang, Bryan Hooi, Yujun Cai, Zhen Xiong, Nanyun Peng and Kai-Wei Chang. 2025. Vulnerability of LLMs to Vertically Aligned Text Manipulations. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

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

```bibtex
@inproceedings{li2025vulnerability,
  title = {{Vulnerability of LLMs to Vertically Aligned Text Manipulations}},
  author = {Zhecheng Li, Yiwei Wang, Bryan Hooi, Yujun Cai, Zhen Xiong, Nanyun Peng and Kai-Wei Chang},
  booktitle = {Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2025},
  doi = {10.18653/v1/2025.acl-long.978},
  url = {https://doi.org/10.18653/v1/2025.acl-long.978}
}
```

- [← previous Uncertainty Distillation: Teaching Language Models to Express Semantic Confidence](https://md.research.elamur.ai/free/en/papers/hager-2025-uncertainty-distillation-teaching-language-models/)
- [next → ZERA: Zero-init Instruction Evolving Refinement Agent – From Zero Instructions to Structured Prompts via Principle-based Optimization](https://md.research.elamur.ai/free/en/papers/yi-2025-zera-zero-init-instruction-evolving/)
- [↑ back to papers of 2025](https://md.research.elamur.ai/free/en/papers/2025/page/11/)

## References

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

1. Tom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, et al. 2020. [Language models are few-shot learners](https://md.research.elamur.ai/free/en/papers/brown-2020-language-models-few-shot-learners/).
2. Patrick Chao, Alexander Robey, Edgar Dobriban, Hamed Hassani, George J. Pappas, and Eric Wong. 2024. [Jailbreaking black box large language models in twenty queries](https://md.research.elamur.ai/free/en/papers/chao-2023-jailbreaking-black-box-large-language/). cjadams, Jeffrey Sorensen, Julia Elliott, Lucas Dixon, Mark McDonald, nithum, and Will Cukierski. 2017. Toxic comment classification ch…
3. Gelei Deng, Yi Liu, Yuekang Li, Kailong Wang, Ying Zhang, Zefeng Li, Haoyu Wang, Tianwei Zhang, and Yang Liu. 2024. Masterkey: Automated jailbreaking of large language model chatbots. In Proceedings 2024 Network and Distributed System Security Symposium, NDSS 2024. Internet Society. [find](https://scholar.google.com/scholar?q=Gelei+Deng%2C+Yi+Liu%2C+Yuekang+Li%2C+Kailong+Wang%2C+Ying+Zhang%2C+Zefeng+Li%2C+Haoyu+Wang%2C+Tianwei+Zhang%2C+and+Yang+Liu%2E+2024%2E+Masterkey%3A+Automated+jailbreaking+of+large+language+model+chatbots%2E+In+Proceedings+2)
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/).
5. Zhichen Dong, Zhanhui Zhou, Chao Yang, Jing Shao, and Yu Qiao. 2024. Attacks, defenses and evaluations for llm conversation safety: A survey. [find](https://scholar.google.com/scholar?q=Zhichen+Dong%2C+Zhanhui+Zhou%2C+Chao+Yang%2C+Jing+Shao%2C+and+Yu+Qiao%2E+2024%2E+Attacks%2C+defenses+and+evaluations+for+llm+conversation+safety%3A+A+survey%2E)
6. Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, Anirudh Goyal, Anthony Hartshorn, Aobo Yang, et al. 2024. [The llama 3 herd of models](https://md.research.elamur.ai/free/en/papers/grattafiori-2024-llama-3-herd-models/).
7. John Fields, Kevin Chovanec, and Praveen Madiraju. 2024a. A survey of text classification with transformers: How wide? how large? how long? how accurate? how expensive? how safe? IEEE Access, 12:6518–6531. [find](https://scholar.google.com/scholar?q=John+Fields%2C+Kevin+Chovanec%2C+and+Praveen+Madiraju%2E+2024a%2E+A+survey+of+text+classification+with+transformers%3A+How+wide%3F+how+large%3F+how+long%3F+how+accurate%3F+how+expensive%3F+how+safe%3F+IEEE+Access%2C+12%3A6518%E2%80%93)
8. John Fields, Kevin Chovanec, and Praveen Madiraju. 2024b. A survey of text classification with transformers: How wide? how large? how long? how accurate? how expensive? how safe? IEEE Access. [find](https://scholar.google.com/scholar?q=John+Fields%2C+Kevin+Chovanec%2C+and+Praveen+Madiraju%2E+2024b%2E+A+survey+of+text+classification+with+transformers%3A+How+wide%3F+how+large%3F+how+long%3F+how+accurate%3F+how+expensive%3F+how+safe%3F+IEEE+Access%2E)
9. Santiago González-Carvajal and Eduardo C Garrido-Merchán. 2020. Comparing bert against traditional machine learning text classification. arXiv preprint arXiv:2005.13012. [find](https://scholar.google.com/scholar?q=Santiago+Gonz%C3%A1lez%2DCarvajal+and+Eduardo+C+Garrido%2DMerch%C3%A1n%2E+2020%2E+Comparing+bert+against+traditional+machine+learning+text+classification%2E+arXiv+preprint+arXiv%3A2005%2E13012%2E)
10. Jeremy Howard and Sebastian Ruder. 2018. Universal language model fine-tuning for text classification. [find](https://scholar.google.com/scholar?q=Jeremy+Howard+and+Sebastian+Ruder%2E+2018%2E+Universal+language+model+fine%2Dtuning+for+text+classification%2E)
11. Zhanhao Hu, Julien Piet, Geng Zhao, Jiantao Jiao, and David Wagner. 2024. Toxicity detection for free. [find](https://scholar.google.com/scholar?q=Zhanhao+Hu%2C+Julien+Piet%2C+Geng+Zhao%2C+Jiantao+Jiao%2C+and+David+Wagner%2E+2024%2E+Toxicity+detection+for+free%2E)
12. Takeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo, and Yusuke Iwasawa. 2023. [Large language models are zero-shot reasoners](https://md.research.elamur.ai/free/en/papers/kojima-2022-large-language-models-zero-shot/).
13. Aristides Milios, Siva Reddy, and Dzmitry Bahdanau. 2023. In-context learning for text classification with many labels. In Proceedings of the 1st GenBench Workshop on (Benchmarking) Generalisation in NLP, pages 173–184, Singapore. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Aristides+Milios%2C+Siva+Reddy%2C+and+Dzmitry+Bahdanau%2E+2023%2E+In%2Dcontext+learning+for+text+classification+with+many+labels%2E+In+Proceedings+of+the+1st+GenBench+Workshop+on+%28Benchmarking%29+Generalisation+in+)
14. Shervin Minaee, Nal Kalchbrenner, Erik Cambria, Narjes Nikzad, Meysam Chenaghlu, and Jianfeng Gao. 2021. Deep learning based text classification: A comprehensive review. [find](https://scholar.google.com/scholar?q=Shervin+Minaee%2C+Nal+Kalchbrenner%2C+Erik+Cambria%2C+Narjes+Nikzad%2C+Meysam+Chenaghlu%2C+and+Jianfeng+Gao%2E+2021%2E+Deep+learning+based+text+classification%3A+A+comprehensive+review%2E)
15. Jonathan Rusert. 2024. Vertattack: Taking advantage of text classifiers’ horizontal vision. In North American Chapter of the Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Jonathan+Rusert%2E+2024%2E+Vertattack%3A+Taking+advantage+of+text+classifiers%E2%80%99+horizontal+vision%2E+In+North+American+Chapter+of+the+Association+for+Computational+Linguistics%2E)
16. Marcin Michał Mirończuk and Jarosław Protasiewicz. 2018. A recent overview of the state-of-the-art elements of text classification. Expert Systems with Applications, 106:36–54. [find](https://scholar.google.com/scholar?q=Marcin+Micha%C5%82+Miron%CC%81czuk+and+Jaros%C5%82aw+Protasiewicz%2E+2018%2E+A+recent+overview+of+the+state%2Dof%2Dthe%2Dart+elements+of+text+classification%2E+Expert+Systems+with+Applications%2C+106%3A36%E2%80%9354%2E)
17. Victor Sanh, Lysandre Debut, Julien Chaumond, and Thomas Wolf. 2020. Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter. [find](https://scholar.google.com/scholar?q=Victor+Sanh%2C+Lysandre+Debut%2C+Julien+Chaumond%2C+and+Thomas+Wolf%2E+2020%2E+Distilbert%2C+a+distilled+version+of+bert%3A+smaller%2C+faster%2C+cheaper+and+lighter%2E)
18. OpenAI, Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, Red Avila, et al. 2024. [Gpt-4 technical report](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/).
19. Long Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul Christiano, Jan Leike, and Ryan Lowe. 2022. [Train…](https://md.research.elamur.ai/free/en/papers/ouyang-2022-training-language-models-follow-instructions/)
20. Bo Pang and Lillian Lee. 2005. [Seeing stars: Exploiting class relationships for sentiment categorization with respect to rating scales](https://md.research.elamur.ai/free/en/papers/pang-2005-seeing-stars-exploiting-class-relationships/). In Proceedings of the 43rd Annual Meeting of the Association for Computational Linguistics (ACL’05), pages 115–124, Ann Arbor, Michigan. Association for Computation…
21. Fábio Perez and Ian Ribeiro. 2022. [Ignore previous prompt: Attack techniques for language models](https://md.research.elamur.ai/free/en/papers/perez-2022-ignore-previous-prompt-attack-techniques/).
22. Nusrat Jahan Prottasha, Abdullah As Sami, Md Kowsher, Saydul Akbar Murad, Anupam Kumar Bairagi, Mehedi Masud, and Mohammed Baz. 2022. Transfer learning for sentiment analysis using bert based supervised fine-tuning. Sensors, 22(11):4157. [find](https://scholar.google.com/scholar?q=Nusrat+Jahan+Prottasha%2C+Abdullah+As+Sami%2C+Md+Kowsher%2C+Saydul+Akbar+Murad%2C+Anupam+Kumar+Bairagi%2C+Mehedi+Masud%2C+and+Mohammed+Baz%2E+2022%2E+Transfer+learning+for+sentiment+analysis+using+bert+based+supervis)
23. Rukhma Qasim, Waqas Haider Bangyal, Mohammed A Alqarni, and Abdulwahab Ali Almazroi. 2022. A fine-tuned bert-based transfer learning approach for text classification. Journal of healthcare engineering, 2022(1):3498123. [find](https://scholar.google.com/scholar?q=Rukhma+Qasim%2C+Waqas+Haider+Bangyal%2C+Mohammed+A+Alqarni%2C+and+Abdulwahab+Ali+Almazroi%2E+2022%2E+A+fine%2Dtuned+bert%2Dbased+transfer+learning+approach+for+text+classification%2E+Journal+of+healthcare+engineering)
24. Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language models are unsupervised multitask learners. OpenAI. Accessed: 2024-11-15. [find](https://scholar.google.com/scholar?q=Alec+Radford%2C+Jeffrey+Wu%2C+Rewon+Child%2C+David+Luan%2C+Dario+Amodei%2C+and+Ilya+Sutskever%2E+2019%2E+Language+models+are+unsupervised+multitask+learners%2E+OpenAI%2E+Accessed%3A+2024%2D11%2D15%2E)
25. Rafael Rafailov, Archit Sharma, Eric Mitchell, Stefano Ermon, Christopher D. Manning, and Chelsea Finn. 2024. [Direct preference optimization: Your language model is secretly a reward model](https://md.research.elamur.ai/free/en/papers/rafailov-2023-direct-preference-optimization-language-model/).
26. Ben Rodrawangpai and Witawat Daungjaiboon. 2022. Improving text classification with transformers and layer normalization. Machine Learning with Applications, 10:100403. [find](https://scholar.google.com/scholar?q=Ben+Rodrawangpai+and+Witawat+Daungjaiboon%2E+2022%2E+Improving+text+classification+with+transformers+and+layer+normalization%2E+Machine+Learning+with+Applications%2C+10%3A100403%2E)
27. Sergio Rojas-Galeano. 2024. Zero-shot spam email classification using pre-trained large language models. arXiv preprint arXiv:2405.15936. — [arXiv:2405.15936](https://arxiv.org/abs/2405.15936) [find](https://scholar.google.com/scholar?q=Sergio+Rojas%2DGaleano%2E+2024%2E+Zero%2Dshot+spam+email+classification+using+pre%2Dtrained+large+language+models%2E+arXiv+preprint+arXiv%3A2405%2E15936%2E+%E2%80%94+%5BarXiv%3A2405%2E15936%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2405%2E15936%29)
28. Melanie Sclar, Yejin Choi, Yulia Tsvetkov, and Alane Suhr. 2024. Quantifying language models’ sensitivity to spurious features in prompt design or: How i learned to start worrying about prompt formatting. [find](https://scholar.google.com/scholar?q=Melanie+Sclar%2C+Yejin+Choi%2C+Yulia+Tsvetkov%2C+and+Alane+Suhr%2E+2024%2E+Quantifying+language+models%E2%80%99+sensitivity+to+spurious+features+in+prompt+design+or%3A+How+i+learned+to+start+worrying+about+prompt+formatt)
29. Richard Socher, Alex Perelygin, Jean Wu, Jason Chuang, Christopher D. Manning, Andrew Ng, and Christopher Potts. 2013. [Recursive deep models for semantic compositionality over a sentiment treebank](https://md.research.elamur.ai/free/en/papers/socher-2013-recursive-deep-models-semantic-compositionality/). In Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, pages 1631–…
30. Chi Sun, Xipeng Qiu, Yige Xu, and Xuanjing Huang. 2020. [How to fine-tune bert for text classification?](https://md.research.elamur.ai/free/en/papers/sun-2019-fine-tune-bert-text-classification/)
31. Gemma Team, Morgane Riviere, Shreya Pathak, Pier Giuseppe Sessa, Cassidy Hardin, Surya Bhupatiraju, Léonard Hussenot, Thomas Mesnard, Bobak Shahriari, Alexandre Ramé, Johan Ferret, Peter Liu, et al. 2024. Gemma 2: Improving open language models at a practical size. [find](https://scholar.google.com/scholar?q=Gemma+Team%2C+Morgane+Riviere%2C+Shreya+Pathak%2C+Pier+Giuseppe+Sessa%2C+Cassidy+Hardin%2C+Surya+Bhupatiraju%2C+L%C3%A9onard+Hussenot%2C+Thomas+Mesnard%2C+Bobak+Shahriari%2C+Alexandre+Ram%C3%A9%2C+Johan+Ferret%2C+Peter+Liu%2C+et+al%2E+2)
32. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, Dan Bikel, Lukas Blecher, et al. 2023. [Llama 2: Open foundation and fine-tuned chat models](https://md.research.elamur.ai/free/en/papers/touvron-2023-llama-2-open-foundation-fine/).
33. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. [Attention is all you need](https://md.research.elamur.ai/free/en/papers/vaswani-2017-attention-all-need/). In Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS’17, page 6000–6010, Red Hook, NY, USA. Cur…
34. Alex Wang, Amanpreet Singh, Julian Michael, Felix Hill, Omer Levy, and Samuel R. Bowman. 2019. [Glue: A multi-task benchmark and analysis platform for natural language understanding](https://md.research.elamur.ai/free/en/papers/wang-2018-glue-multi-task-benchmark-analysis/).
35. Cindy Wang and Michele Banko. 2021. Practical transformer-based multilingual text classification. In North American Chapter of the Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Cindy+Wang+and+Michele+Banko%2E+2021%2E+Practical+transformer%2Dbased+multilingual+text+classification%2E+In+North+American+Chapter+of+the+Association+for+Computational+Linguistics%2E)
36. Yau-Shian Wang and Yingshan Chang. 2022. Toxicity detection with generative prompt-based inference. [find](https://scholar.google.com/scholar?q=Yau%2DShian+Wang+and+Yingshan+Chang%2E+2022%2E+Toxicity+detection+with+generative+prompt%2Dbased+inference%2E)
37. Yiwei Wang, Yujun Cai, Muhao Chen, Yuxuan Liang, and Bryan Hooi. 2023. Primacy effect of ChatGPT. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 108–115, Singapore. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Yiwei+Wang%2C+Yujun+Cai%2C+Muhao+Chen%2C+Yuxuan+Liang%2C+and+Bryan+Hooi%2E+2023%2E+Primacy+effect+of+ChatGPT%2E+In+Proceedings+of+the+2023+Conference+on+Empirical+Methods+in+Natural+Language+Processing%2C+pages+108%E2%80%931)
38. Alex Warstadt, Amanpreet Singh, and Samuel R. Bowman. 2019. Neural network acceptability judgments. [find](https://scholar.google.com/scholar?q=Alex+Warstadt%2C+Amanpreet+Singh%2C+and+Samuel+R%2E+Bowman%2E+2019%2E+Neural+network+acceptability+judgments%2E)
39. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. 2023. [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/).
40. Jason Wei and Kai Zou. 2019. [Eda: Easy data augmentation techniques for boosting performance on text classification tasks](https://md.research.elamur.ai/free/en/papers/wei-2019-eda-easy-data-augmentation-techniques/).
41. An Yang, Baosong Yang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Zhou, Chengpeng Li, Chengyuan Li, Dayiheng Liu, Fei Huang, Guanting Dong, Haoran Wei, Huan Lin, Jialong Tang, et al. 2024. Qwen2 technical report. [find](https://scholar.google.com/scholar?q=An+Yang%2C+Baosong+Yang%2C+Binyuan+Hui%2C+Bo+Zheng%2C+Bowen+Yu%2C+Chang+Zhou%2C+Chengpeng+Li%2C+Chengyuan+Li%2C+Dayiheng+Liu%2C+Fei+Huang%2C+Guanting+Dong%2C+Haoran+Wei%2C+Huan+Lin%2C+Jialong+Tang%2C+et+al%2E+2024%2E+Qwen2+technical)
42. Wenpeng Yin, Jamaal Hay, and Dan Roth. 2019. [Benchmarking zero-shot text classification: Datasets, evaluation and entailment approach](https://md.research.elamur.ai/free/en/papers/yin-2019-benchmarking-zero-shot-text-classification/).

*Updated 30 October 2026*