--- title: "Xiangyu Qi — author profile and papers · Bias, fairness & safety" description: "Xiangyu Qi, Jiangsu Academy of Agricultural Sciences: 5 papers on Elamur Research (2023–2024), h-index 17 — affiliations, topics…" canonical: "https://research.elamur.ai/free/en/authors/xiangyu-qi/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/xiangyu-qi/) [home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / Xiangyu Qi # Xiangyu Qi Xiangyu Qi has 5 papers in our library, in 2023–2024, most of them on [Bias, fairness & safety](https://md.research.elamur.ai/free/en/topics/bias-fairness-safety/), [Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/); often with [Peter Henderson](https://md.research.elamur.ai/free/en/authors/peter-henderson/), [Prateek Mittal](https://md.research.elamur.ai/free/en/authors/prateek-mittal/), [Kaixuan Huang](https://md.research.elamur.ai/free/en/authors/kaixuan-huang/). 5 papers in our library · 2023–2024 [OpenAlex](https://openalex.org/A5036739334) · [ORCID](https://orcid.org/0009-0005-9842-1093) · [**h-index 17**](https://openalex.org/authors/A5036739334) · [77 works](https://openalex.org/works?filter=author.id:A5036739334) · [963 citations](https://openalex.org/works?filter=author.id:A5036739334&sort=cited_by_count:desc) OpenAlex, 2026-10-14 [📄 5 works in our library](https://md.research.elamur.ai/free/en/authors/xiangyu-qi/#papers) · [🔗 Cited 905 times in our library](https://md.research.elamur.ai/free/en/authors/xiangyu-qi/cited-by/) · [2 more in our library without a page yet](#also) · [⭐ Paper Stars ranking: #3034](https://md.research.elamur.ai/free/en/paper-stars/page/31/#star-3034) The profile is the library's match of the name; two people with one name are not always told apart. Elsewhere: [Google Scholar](https://scholar.google.com/citations?view_op=search_authors&mauthors=Xiangyu+Qi) · [Semantic Scholar](https://www.semanticscholar.org/search?q=Xiangyu+Qi) · [dblp](https://dblp.org/search?q=Xiangyu+Qi) · [ACL Anthology](https://aclanthology.org/search/?q=Xiangyu+Qi) ## Affiliations [🏛️**Jiangsu Academy of Agricultural Sciences** Nanjing · China 2020–2026 · 25 papers*current*](https://ror.org/001f9e125) · [🏥**Shandong Provincial Hospital** Jinan · China 2020–2026 · 10 papers*current*](https://ror.org/02ar2nf05) · [🎓**Shandong First Medical University** Jinan · China 2022–2026 · 6 papers*current*](https://md.research.elamur.ai/free/en/universities/shandong-first-medical-university/) · [🎓**Northwest A&F University** Yangling · China 2022–2026 · 3 papersEmployment since 2022 · 学生, 草业与草原学院*current*](https://ror.org/0051rme32) · [🎓**Princeton University** Princeton · United States 2022–2025 · 15 papers](https://md.research.elamur.ai/free/en/universities/princeton-university/) · [🎓**Nanjing Agricultural University** Nanjing · China 2013–2022 · 14 papers](https://ror.org/05td3s095) · [🎓**Shandong University** Jinan · China 2020–2023 · 7 papers](https://md.research.elamur.ai/free/en/universities/shandong-university/) · [🏛️**Ministry of Agriculture and Rural Affairs** Beijing · China 2021–2022 · 3 papers](https://ror.org/05ckt8b96) · [🎓**Beijing University of Agriculture** Beijing · China 2022 · 2 papers](https://ror.org/03t9adt98) From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 3 more institutions on single bylines are not listed. ## Fields, topics and concepts Of its works in our library, with the number of works on each. 📚 Shelves 7 The field of science this work belongs to (OpenAlex fields).[Computer Science · 5 works](https://md.research.elamur.ai/free/en/shelves/computer-science/) · [Artificial Intelligence · 5 works](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · [Safety Research · 4 works](https://md.research.elamur.ai/free/en/shelves/social-sciences/safety-research/) · [Social Sciences · 4 works](https://md.research.elamur.ai/free/en/shelves/social-sciences/) · [Arts and Humanities · 2 works](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · [Language and Linguistics · 2 works](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · [Computer Vision and Pattern Recognition · 1 work](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) 🧭 Topics 2 Research subjects and languages the work is about.[Bias, fairness & safety · 5 works](https://md.research.elamur.ai/free/en/topics/bias-fairness-safety/) · [Language models & pretraining · 3 works](https://md.research.elamur.ai/free/en/topics/language-models/) ## Works by year ### 2020s [**2024** 3 papers](https://md.research.elamur.ai/free/en/authors/xiangyu-qi/year/2024/) · [**2023** 2 papers](https://md.research.elamur.ai/free/en/authors/xiangyu-qi/year/2023/) ## Papers The most cited first (OpenAlex and our library). 1. **[Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!](https://md.research.elamur.ai/free/en/papers/qi-2023-fine-tuning-aligned-language-models/)** *(2023 · 355 citations)* Current safety alignment methods, such as instruction tuning and RLHF, constrain how a model behaves at inference time. 2. **[Visual Adversarial Examples Jailbreak Aligned Large Language Models](https://md.research.elamur.ai/free/en/papers/qi-2023-visual-adversarial-examples-jailbreak-aligned/)** *(2023 · 240 citations)* Adding image inputs to large language models creates new security risks. The authors argue that images are continuous and high-dimensional, which makes them much easier to manipulate adversarially than discrete text. Because these models can do so many things, a successful attack can do more than cause a… 3. **[Safety Alignment Should Be Made More Than Just a Few Tokens Deep](https://md.research.elamur.ai/free/en/papers/qi-2024-safety-alignment-should-made-more/)** *(2024 · 97 citations)* The paper asks why safety-aligned LLMs are so easy to break, whether through adversarial suffixes, prefilling, tweaked decoding parameters, or a few steps of fine-tuning. 4. **[SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal](https://md.research.elamur.ai/free/en/papers/xie-2024-sorry-bench-systematically-evaluating-large/)** *(2024 · 87 citations)* SORRY-Bench is a benchmark that measures how reliably aligned LLMs decline unsafe requests. The authors identify three weaknesses in earlier safety evaluations. Their category schemes are coarse and inconsistent, and they concentrate on a few topics such as fraud while barely testing others like self-harm. They… 5. **[Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications](https://md.research.elamur.ai/free/en/papers/wei-2024-assessing-brittleness-safety-alignment-pruning/)** *(2024 · 66 citations)* Aligned LLMs can be jailbroken or lose their safeguards even after harmless fine-tuning. The authors ask whether this fragility can be traced to specific parts of the weights. Using the Llama2-chat family, they attribute behaviour to individual neurons and to low-rank components of weight matrices. They use two… ## All works 7 works of theirs in our library, the most cited first; 5 with a page here, the others link their original until we summarise them. 1. [**Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To! arXiv:2310.03693**Xiangyu Qi, Yi Zeng, Tinghao Xie, et al.*2023 · 355 citations*](https://md.research.elamur.ai/free/en/papers/qi-2023-fine-tuning-aligned-language-models/) 2. **[Visual Adversarial Examples Jailbreak Aligned Large Language Models](https://md.research.elamur.ai/free/en/papers/qi-2023-visual-adversarial-examples-jailbreak-aligned/) [Proceedings of the AAAI Conference on Artificial Intelligence](https://md.research.elamur.ai/free/en/venues/proceedings-of-the-aaai-conference-on-artificial-intelligence/)**Xiangyu Qi, Kaixuan Huang, Ashwinee Panda, et al.*2023 · 240 citations* 3. [**Safety Alignment Should Be Made More Than Just a Few Tokens Deep arXiv:2406.05946**Xiangyu Qi, Ashwinee Panda, Kaifeng Lyu, et al.*2024 · 97 citations*](https://md.research.elamur.ai/free/en/papers/qi-2024-safety-alignment-should-made-more/) 4. [**SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal arXiv:2406.14598**Tinghao Xie, Xiangyu Qi, Yi Zeng, et al.*2024 · 87 citations*](https://md.research.elamur.ai/free/en/papers/xie-2024-sorry-bench-systematically-evaluating-large/) 5. [**Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications arXiv:2402.05162**Boyi Wei, Kaixuan Huang, Yangsibo Huang, et al.*2024 · 66 citations*](https://md.research.elamur.ai/free/en/papers/wei-2024-assessing-brittleness-safety-alignment-pruning/) 6. **[AI Risk Management Should Incorporate Both Safety and Security ↗](https://arxiv.org/abs/2405.19524) arXiv (Cornell University) · no summary yet**Xiangyu Qi et al.*2024 · 3 citations* 7. **[The Model Hears You: Audio Language Model Deployments Should Consider the Principle of Least Privilege ↗](https://doi.org/10.1609/aies.v8i2.36625) Proceedings of the AAAI/ACM Conference on AI Ethics and Society · no summary yet**Luxi He et al.*2025 · 1 citation* ## Co-authors here 12 of 17 co-authors with a page, the most papers together first. - [Peter Henderson](https://md.research.elamur.ai/free/en/authors/peter-henderson/) ×4 - [Prateek Mittal](https://md.research.elamur.ai/free/en/authors/prateek-mittal/) ×4 - [Kaixuan Huang](https://md.research.elamur.ai/free/en/authors/kaixuan-huang/) ×3 - [Tinghao Xie](https://md.research.elamur.ai/free/en/authors/tinghao-xie/) ×3 - [Ashwinee Panda](https://md.research.elamur.ai/free/en/authors/ashwinee-panda/) ×2 - [Boyi Wei](https://md.research.elamur.ai/free/en/authors/boyi-wei/) ×2 - [Mengdi Wang](https://md.research.elamur.ai/free/en/authors/mengdi-wang/) ×2 - [Ruoxi Jia](https://md.research.elamur.ai/free/en/authors/ruoxi-jia/) ×2 - [Yangsibo Huang](https://md.research.elamur.ai/free/en/authors/yangsibo-huang/) ×2 - [Yi Zeng](https://md.research.elamur.ai/free/en/authors/yi-zeng/) ×2 - [Bo Li](https://md.research.elamur.ai/free/en/authors/bo-li/) - [Dacheng Li](https://md.research.elamur.ai/free/en/authors/dacheng-li/) ## Venues Where their works here were published, the most works first, with the editions of those that have a page. - [Proceedings of the AAAI Conference on Artificial Intelligence](https://md.research.elamur.ai/free/en/venues/proceedings-of-the-aaai-conference-on-artificial-intelligence/): [Proceedings of the AAAI Conference on Artificial Intelligence 2023](https://md.research.elamur.ai/free/en/venues/proceedings-of-the-aaai-conference-on-artificial-intelligence/2023/) - Proceedings of the AAAI/ACM Conference on AI Ethics and Society ## Cited by Papers of our library cite their works 905 times (of the 149,439 papers whose bibliographies we read); [215 citing papers with a page here →](https://md.research.elamur.ai/free/en/authors/xiangyu-qi/cited-by/). Updated 2 November 2026