Pre-trained MoE models exhibit deep-layer routing collapse for low-resource languages like Hebrew, largely corrected by continual pre-training on balanced bilingual data, with consistent patterns observed in Japanese.
Title resolution pending
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
Liberata introduces a graph-based system using continuous contribution shares and weighted citations to derive metrics for impact, risk, collaboration, and quality control in academic publishing.
Random feature selection outperforms many state-of-the-art unsupervised feature selection methods in both performance and efficiency.
citing papers explorer
-
Mixture of Experts for Low-Resource LLMs
Pre-trained MoE models exhibit deep-layer routing collapse for low-resource languages like Hebrew, largely corrected by continual pre-training on balanced bilingual data, with consistent patterns observed in Japanese.
-
Liberata -- Graph Scientometrics for a Share Based System of Academic Publishing
Liberata introduces a graph-based system using continuous contribution shares and weighted citations to derive metrics for impact, risk, collaboration, and quality control in academic publishing.
-
Worse than Random: The Importance of a Baseline for Unsupervised Feature Selection
Random feature selection outperforms many state-of-the-art unsupervised feature selection methods in both performance and efficiency.