Hanxu Hu
Scholar

Hanxu Hu

Google Scholar ID: r9fCUd8AAAAJ
University of Zurich, University of Edinburgh
Large Language ModelsMachine Learning
Citations & Impact
All-time
Citations
145
 
H-index
5
 
i10-index
4
 
Publications
12
 
Co-authors
0
 
Resume (English only)
Academic Achievements
  • Papers published:
  • - QueST: Incentivizing LLMs to Generate Difficult Problems (Preprint)
  • - Source-primed Multi-turn Conversation Helps LLMs Translate Documents (EMNLP 2025 Findings)
  • - Fine-tuning Large Language Models with Sequential Instructions (NAACL 2025 Main)
  • - Chain-of-Symbol Prompting For Spatial Reasoning in Large Language Models (COLM 2024)
  • - Meta-Learning For Multi-Modal Cross-lingual Transfer (MRL at EMNLP 2023, Best Paper Nomination)
  • Workshops co-organized:
  • - Workshop on Multilingual and Multicultural Evaluation (EACL 2026)
  • - Workshop on Multi-Turn Interactions in LLMs (NeurIPS 2025)
Research Experience
  • Was a research intern at Microsoft Research mentored by Dr. Xingxing Zhang; was a member of EdinburghNLP advised by Prof. Edoardo Ponti; and a research assistant at WestlakeNLP advised by Prof. Yue Zhang.
Education
  • Started a PhD at the University of Zurich in Fall 2024, co-supervised by Prof. Rico Sennrich from the University of Zurich and Prof. Ivan Titov from the University of Edinburgh, and will work closely with Dr. Jannis Vamvas.
Background
  • Research interests include the use of synthetic data for reasoning and post-training of Large Language Models. Previously, also worked on multilinguality.
Miscellany
  • Personal interests: Read 'The Bitter Lesson' and learned a lot from it. Has a cat named Derby, born in December 2024.
Co-authors
0 total
Co-authors: 0 (list not available)