Showing stat.MLShow all
2 papers · 1 filter
stat.ML2025
Spectral Thresholds for Identifiability and Stability:Finite-Sample Phase Transitions in High-Dimensional Learning
William Hao-Cheng Huang
In high-dimensional learning, models remain stable until they collapse abruptly once the sample size falls below a critical level. This instability is not algorithm-specific but a…
stat.ML2025
Spectral Identifiability for Interpretable Probe Geometry
William Hao-Cheng Huang
Linear probes are widely used to interpret and evaluate neural representations, yet their reliability remains unclear, as probes may appear accurate in some regimes but collapse un…