Xinyuan Cheng 程心远

Master’s Student in Computational Linguistics
LMU Munich, Germany
GitHub · Email · Linkeln

---

🧠 Research Interests

• Large Language Models (LLMs)
• Interpretability and Evaluation of NLP Models
• Vision-Language Alignment
• Simplification and Summarization

---

🎓 Education

M.A. Computational Linguistics, Ludwig-Maximilians-Universität München
2024 – 2026 (expected)
Courses: Machine Learning, Data Mining, NLP, Summarization, Information Retrieval

B.A. German Studies, Shanghai Jiao Tong University
2019 – 2023
Thesis: Linguistic Ambiguity in German and its Interpretation by LLMs

---

📄 Selected Projects

• Meta-Evaluation of RAG Metrics
  Evaluated robustness of reference-free metrics (e.g., RAGAs, SummaC) across prompt/model variations for retrieval-augmented generation.
  Work in progress

• SWiPe Dataset Reanalysis
  Studied document-level simplification metrics (SARI, LENS-SALSA, QuestEval); compared original and reversed edits.
  GitHub Repo)

---

🛠️ Tools & Skills

• Python (NumPy, HuggingFace, fastText, PyTorch)
• Git, Bash, Linux, LaTeX
• English (fluent), German (advanced), Chinese (native)

---

📬 Contact

Email: Xinyuan.Cheng@campus.lmu.de
GitHub: noah-13
Location: Munich, Germany