A silicon-native modular system with parallel live distillation and a tight-bottleneck autoencoder achieves parameter isolation, autonomous task discovery, and strong retention across vision and language tasks without a student fidelity gap.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Modular Continual Learning via Zero-Leakage Reconstruction Routing and Autonomous Task Discovery
A silicon-native modular system with parallel live distillation and a tight-bottleneck autoencoder achieves parameter isolation, autonomous task discovery, and strong retention across vision and language tasks without a student fidelity gap.