OSDN adds online diagonal preconditioning to the Delta Rule, preserving chunkwise parallelism while proving super-geometric convergence and delivering 32-39% recall gains at 340M-1.3B scales.
Adaptive subgradient methods for online learning and stochastic optimization.Journal of Machine Learning Research, 12:2121–2159
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
years
2026 2verdicts
UNVERDICTED 2roles
method 1polarities
use method 1representative citing papers
LBW-Guard is a bounded autonomous control layer above AdamW that improves stability, reduces perplexity, and speeds up training for Qwen2.5 models under learning-rate stress on WikiText-103.
citing papers explorer
-
OSDN: Improving Delta Rule with Provable Online Preconditioning in Linear Attention
OSDN adds online diagonal preconditioning to the Delta Rule, preserving chunkwise parallelism while proving super-geometric convergence and delivering 32-39% recall gains at 340M-1.3B scales.
-
Learn-by-Wire Training Control Governance: Bounded Autonomous Training Under Stress for Stability and Efficiency
LBW-Guard is a bounded autonomous control layer above AdamW that improves stability, reduces perplexity, and speeds up training for Qwen2.5 models under learning-rate stress on WikiText-103.