Tianyi Hu

Tianyi Hu

PhD Student

Aarhus University

Research Interests

Natural Language Processing
Information Retrieval
Computational Social Science

About

I am a PhD student in the CLAN Lab, Department of Computer Science at Aarhus University, Denmark, under the supervision of Akhil Arora. I am also affiliated with the Danish Pioneer Centre for AI (P1).

Prior to this, I obtained a Master degree in Computer Science from the University of Copenhagen, advised by Daniel Hershcovich and Maria Maistro. I was a visiting student at National Taiwan University, Taiwan in 2024 and a Researcher & Developer in Baidu Search during 2018-2022, and I received my BSc in Software Engineering at Dalian University of Technology, China in 2019.

My current research focuses on human-centric NLP, with a particular interest in understanding the human dimensions of information seeking beyond conventional benchmarks. My recent work explores diversity and cross-cultural variation in information seeking, alongside ways to diagnose potential failure modes in existing systems. More broadly, my research interests include information retrieval and RAG, cultural analysis in NLP, fairness and bias, reasoning, and agents.

I would be happy to explore opportunities for collaboration and supervising student theses/projects. Please feel free to email me to stay in touch.

Selected Publications

View All →

CulturalRiddles: A Multicultural Benchmark of Riddles

My Chiffon Nguyen, Tianyi Hu (co-first author), Vinod Anbalagan, Henry Gagnier, Pouya Sadeghi, …, Marzieh Fadaee, Akhil Arora, Julia Kreutzer, Károly Boczka

Under submission

Where and Why Search-Augmented LLMs Fail When Reasoning over Rapidly Evolving Knowledge

Tianyi Hu, Bhavyajeet Singh, Niket Tandon, Akhil Arora

Under submission

DIVERGE: Diversity-Enhanced RAG for Open-Ended Information Seeking

Tianyi Hu, Niket Tandon, Akhil Arora

Under submission

Oral presentation: The 1st Workshop on Stereotypes Across Cultures in Language Technologies (ACL 2026).

Culinary Crossroads: A RAG Framework for Enhancing Diversity in Cross-Cultural Recipe Adaptation

Tianyi Hu, Andrea Morales-Garzón, Jingyi Zheng, Maria Maistro, Daniel Hershcovich

ACL 2026 (Oral, top 4.15%)

CHASM: Unveiling Covert Advertisements on Chinese Social Media

Jingyi Zheng, Tianyi Hu (co-first author), Yule Liu, Zhen Sun, Zongmin Zhang, Wenhan Dong, Zifan Peng, Xinlei He

NeurIPS 2025

CL-Attack: Textual Backdoor Attacks via Cross-Lingual Triggers

Jingyi Zheng, Tianyi Hu (co-first author), Tianshuo Cong, Xinlei He

AAAI 2025

Less Noise, More Voice: Reinforcement Learning for Reasoning via Instruction Purification

Yiju Guo, Tianyi Hu, Zexu Sun, Yankai Lin

Findings of ACL 2026

TransMeme: A Multi-Agent Framework for Cross-Cultural Meme Transcreation

Jingyi Zheng, Yule Liu, Zifan Peng, Tianyi Hu, Yuemeng Zhao, Xinhu Zheng, Xinlei He

ACM International Conference on Multimedia (ACM MM) 2026

Do LLMs Understand Wine Descriptors Across Cultures? A Benchmark for Cultural Adaptations of Wine Reviews

Chenye Zou, Xingyue Wen, Tianyi Hu, Qian Janice Wang, Daniel Hershcovich

Findings of EMNLP 2025

News

2026-07

Our paper TransMeme has been accepted to ACM Multimedia (ACM MM) 2026!

2026-04

Two papers accepted to ACL 2026!

2025-09

A paper accepted to NeurIPS 2025!

2025-09

A paper accepted to EMNLP 2025!

2025-09

I am excited to officially announce that I will start my PhD at CLAN Lab, Department of Computer Science at Aarhus University.

2025-06

I have graduated from the University of Copenhagen with an excellent GPA (11.7/12.0) and highest possible grade for my Master’s thesis.

2024-12

One paper accepted to AAAI 2025.

2024-10

One paper accepted to EMNLP 2024. And I am awarded Diversity and Inclusion Award of EMNLP 2024.

2024-09

I am granted the Demark Governmental Scholarship for outstanding academic performance.

2024-09

I start exchange study at National Taiwan University!