← Search

Shijie Ren

2 accepted papers

2026

C$^{2}$R: Cross-sample Consistency Regularization Mitigates Feature Splitting and Absorption in Sparse Autoencoders

ICML 2026poster

Sparse Autoencoders (SAEs) are widely used to interpret large language models by decomposing activations into sparse, human-understandable features, but scaling to large dictionaries exposes fundamental challenges. Systematic studies reveal pervasive feature splitting that fragments coherent concept…

Cited by 0SourceScholar
2026

ProtoTS: Learning Hierarchical Prototypes for Explainable Time Series Forecasting

ICLR 2026poster

While deep learning has achieved impressive performance in time series forecasting, it becomes increasingly crucial to understand its decision-making process for building trust in high-stakes scenarios. Existing interpretable models often provide only local and partial explanations, lacking the capa…

Cited by 0SourceScholar