Hanxu Hu

PhD student · University of Zurich

I'm a second-year PhD student at the University of Zurich, co-supervised by Rico Sennrich and Ivan Titov (Edinburgh), working closely with Jannis Vamvas.

My research focuses on reasoning and agentic training of large language models.

Previously: research intern at Microsoft Research with Xingxing Zhang; member of EdinburghNLP with Edoardo Ponti; research assistant at WestlakeNLP with Yue Zhang.

Hanxu with his cat Derby

News

May 2026

Started as a student researcher at Meituan M17 · LongCat, working on long-horizon tasks for agentic training.

Oct 2025

QueST is out on arXiv — incentivizing LLMs to generate difficult problems for code reasoning.

Oct 2025

Co-organizing a workshop on Multilingual and Multicultural Evaluation at EACL 2026 — see you in Morocco.

Oct 2025

Co-organizing the workshop on Multi-Turn Interactions in LLMs at NeurIPS 2025 — see you in San Diego.

Aug 2025

Three papers accepted at EMNLP 2025: multi-turn document translation, multilingual evaluation, and decoding for contamination detection.

Apr 2025

Started as a student researcher at Microsoft Research (Apr–Sep 2025) with Xingxing Zhang.

Mar 2025

Sequential Instruction Tuning (SIT) accepted at NAACL 2025.

Sep 2024

Started my PhD at the University of Zurich.

Selected Papers

Full list on Google Scholar.

QueST: Incentivizing LLMs to Generate Difficult Problems

Hanxu Hu, Xingxing Zhang, Jannis Vamvas, Rico Sennrich, Furu Wei

Preprint, 2025

Source-Primed Multi-Turn Conversation Helps LLMs Translate Documents

Hanxu Hu, Jannis Vamvas, Rico Sennrich

EMNLP 2025 Findings

Fine-tuning Large Language Models with Sequential Instructions

Hanxu Hu*, Simon Yu*, Pinzhen Chen*, Edoardo M. Ponti

NAACL 2025 (Main)

Chain-of-Symbol Prompting for Spatial Reasoning in LLMs

Hanxu Hu*, Hongyuan Lu*, Huajian Zhang, Yun-Ze Song, Yue Zhang

COLM 2024

Meta-Learning for Multi-Modal Cross-Lingual Transfer

Hanxu Hu, Frank Keller

MRL @ EMNLP 2023Best Paper Nomination

Workshops Organized

NeurIPS '25 Multi-Turn Interactions in LLMs — co-organizer
EACL '26 Multilingual and Multicultural Evaluation — co-organizer

Service & Teaching

Reviewer EMNLP 2023, NeurIPS 2024, NAACL 2025, ACL 2025, ICML 2025, COLM 2025
TA Machine Learning Practical (Edinburgh, 23–24); Large Language Models (Zurich, 24 Fall)

Experience

2026 – Student Researcher, Meituan M17 · LongCat — long-horizon tasks for agentic training
2025 Student Researcher, Microsoft Research — code reasoning with Xingxing Zhang
2022–23 Research Assistant, WestlakeNLP — with Yue Zhang