TokAlign++ learns token alignments between LLM vocabularies from monolingual representations to enable faster adaptation, better text compression, and effective token-level distillation across 15 languages with minimal steps.
OFA : A Framework of Initializing Unseen Subword Embeddings for Efficient Large-scale Multilingual Continued Pretraining
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CL 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Interpretability-based selection of vocabulary items plus FragMend initialization reduces token over-fragmentation and improves performance for non-Latin script languages by roughly 20 points over baselines.
citing papers explorer
-
TokAlign++: Advancing Vocabulary Adaptation via Better Token Alignment
TokAlign++ learns token alignments between LLM vocabularies from monolingual representations to enable faster adaptation, better text compression, and effective token-level distillation across 15 languages with minimal steps.
-
Defragmenting Language Models: An Interpretability-based Approach for Vocabulary Expansion
Interpretability-based selection of vocabulary items plus FragMend initialization reduces token over-fragmentation and improves performance for non-Latin script languages by roughly 20 points over baselines.