Zhongyuan Liang

2 accepted papers

2026

Falsifying Sparse Autoencoder Reasoning Features in Language Models

George Ma, Zhongyuan Liang, Irene Y. Chen, Somayeh Sojoudi

ICML 2026poster

We study how reliably sparse autoencoders (SAEs) support claims about reasoning-related internal features in large language models. We first give a stylized analysis showing that sparsity-regularized decoding can preferentially retain stable low-dimensional correlates while suppressing high-dimensio…

Cited by 0 Source ↗Scholar ↗

2023

A Learning Based Hypothesis Test for Harmful Covariate Shift

Tom Ginsberg, Zhongyuan Liang, Rahul G Krishnan

ICLR 2023poster

The ability to quickly and accurately identify covariate shift at test time is a critical and often overlooked component of safe machine learning systems deployed in high-risk domains. While methods exist for detecting when predictions should not be made on out-of-distribution test examples, identif…

Cited by 14 Source ↗PDF ↗Scholar ↗