Zhaoxin Feng (Betty)

Hello there

I am a second-year PhD student at The Hong Kong Polytechnic University, advised by Prof. Emmanuele Chersoni and co-advised by Prof. Huang Chu-Ren. I received my B.Sc. in Educational Technology (minor in Data Science) from Beijing Normal University in 2024.

My research asks two connected questions: do LLMs truly understand human language, and can we rely on them when they interact with humans?

In practice, my work splits into two threads. First, I probe how LLMs represent human language: whether enabling bidirectional attention in autoregressive LLMs reshapes their lexical semantic representations, and how their embeddings compare to human mental lexicon judgments. Second, I study the reliability of LLMs and LLM-based agents when interacting with humans, from sycophancy that chain-of-thought reasoning both mitigates and masks, to memory agents that can recall user preferences but fail to act on them, to reasoning failures on Chinese phonological ambiguities, etc.

Currently

• Apr 2026 Heading to ACL 2026 in San Diego Presenting Good Arguments Against the People Pleasers — on how reasoning both mitigates and masks LLM sycophancy.