SimReg regularization accelerates LLM pretraining convergence by over 30% and raises average zero-shot performance by over 1% across benchmarks.
Joint selection for large-scale pre-training data via policy gradient-based mask learning.arXiv preprint arXiv:2512.24265
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
SimReg: Achieving Higher Performance in the Pretraining via Embedding Similarity Regularization
SimReg regularization accelerates LLM pretraining convergence by over 30% and raises average zero-shot performance by over 1% across benchmarks.