Three last-layer coefficients derived from gradient interference sources predict the forgetting rank order of past classes in rehearsal-based CIL, with self-induced interference as the strongest predictor.
Ramasesh, Ethan Dyer, and Maithra Raghu
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
representative citing papers
Sequential training of multiple tasks followed by unsupervised sleep-like replay partially restores performance across all previously learned tasks in neural networks.
Ranked preference modeling outperforms imitation learning for language model alignment and scales more favorably with model size.
HASGO combines harmony search with universal MLIPs and multi-head replay fine-tuning to locate operando surface reconstructions, demonstrated by identifying the square-pyramidal O5 subsurface motif on Ag(100) during ethylene epoxidation.
citing papers explorer
-
Understanding Imbalanced Forgetting in Rehearsal-Based Class-Incremental Learning
Three last-layer coefficients derived from gradient interference sources predict the forgetting rank order of past classes in rehearsal-based CIL, with self-induced interference as the strongest predictor.
-
Not Just After One: Sleep-Inspired Replay Prevents Catastrophic Forgetting After Sequential Tasks
Sequential training of multiple tasks followed by unsupervised sleep-like replay partially restores performance across all previously learned tasks in neural networks.
-
A General Language Assistant as a Laboratory for Alignment
Ranked preference modeling outperforms imitation learning for language model alignment and scales more favorably with model size.
-
Scalable Prediction of Complex Surface Reconstructions under Operating Conditions via Harmony-Search-Based Global Optimization
HASGO combines harmony search with universal MLIPs and multi-head replay fine-tuning to locate operando surface reconstructions, demonstrated by identifying the square-pyramidal O5 subsurface motif on Ag(100) during ethylene epoxidation.