3 papers
cs.CL2026
Data Scaling as Progressive Coverage of a Predictive Contribution Spectrum
Zihui Song, Shihao Ji, Hongxi Li +2
We investigate the hypothesis that real-data scaling laws are governed by progressive coverage of a latent predictive contribution spectrum rather than by token-frequency tails alo…
cs.LG2026
Task-Driven Kernel Flows: Label Rank Compression and Laplacian Spectral Filtering
Hongxi Li, Chunlin Huang
We present a theory of feature learning in wide L2-regularized networks showing that supervised learning is inherently compressive. We derive a kernel ODE that predicts a "water-fi…
cs.CV2025
DPFormer: Dynamic Prompt Transformer for Continual Learning
Sheng-Kai Huang, Jiun-Feng Chang, Chun-Rong Huang
In continual learning, solving the catastrophic forgetting problem may make the models fall into the stability-plasticity dilemma. Moreover, inter-task confusion will also occur du…