3 papers
cs.CV2026
Rethink MAE with Linear Time-Invariant Dynamics
Zice Wang
Standard representation probing for visual models relies on mathematically permutation-invariant operations like Global Average Pooling (GAP) or CLS tokens, treating patch represen…
cs.CL2026
Framing Effects in Independent-Agent Large Language Models: A Cross-Family Behavioral Analysis
Zice Wang, Zhenyu Zhang
In many real-world applications, large language models (LLMs) operate as independent agents without interaction, thereby limiting coordination. In this setting, we examine how prom…
cs.LG2026
The Malignant Tail: Spectral Segregation of Label Noise in Over-Parameterized Networks
Zice Wang
While implicit regularization facilitates benign overfitting in low-noise regimes, recent theoretical work predicts a sharp phase transition to harmful overfitting as the noise-to-…