About Me

Hello! I’m a first-year Master’s student in Computer Science at Mcgill University and Quebec AI Institute (Mila), advised by Prof. Jackie Chi Kit Cheung. Before my master’s studies, I completed my Bachelor’s degree in Joint Honours Mathematics and Computer Science at McGill University. My research interests center around trustworthy AI

1. Reliability: Abstractions in Reasoning
   • How do large language models (LLMs) form and use abstractions?
   • How do these abstractions interact with generalization and memorization?
2. Robustness: Hallucinations
   • What systematic patterns exist in hallucinations?
   • How can we attribute and mitigate these failure modes in LLMs to make models more robust?

Previously, I participated in the Google CS research mentorship program. In Fall 2024, I served as a Technical Project Manager at McGill Artificial Intelligence Society (MAIS).

Publications and Manuscripts

• Stochastic Parakeets, Peacocks, and Penguins: Irrelevant Context Hallucinations Reveal Class-Based (Mis)Generalization
  In submission to ACL 2025
  Ziling Cheng*, Meng Cao*, Marc-Antoine Rondeau, and Jackie Chi Kit Cheung

• McGill BabyLM Shared Task Submission: The Effects of Data Formatting and Structural Biases.
  In Proceedings of the BabyLM Challenge at CoNLL 2023
  Ziling Cheng, Rahul Aralikatte, Ian Porada, Cesare Spinoso-Di Piano, and Jackie Chi Kit Cheung

• Vārta: A Large-Scale Headline-Generation Dataset for Indic Languages.
  In Findings of ACL 2023
  Rahul Aralikatte*, Ziling Cheng*, Sumanth Doddapaneni, and Jackie Chi Kit Cheung