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.
Mini-Model Adaptation: Efficiently Extending Pretrained Models to New Languages via Aligned Shallow Training
2 Pith papers cite this work, alongside 4 external citations. Polarity classification is still indexing.
2
Pith papers citing it
4
external citations · OpenAlex
fields
cs.CL 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Replacing tokens, freezing the corresponding embeddings, and tuning the rest of the model improves NLU performance on low-resource languages compared to full fine-tuning.
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.
-
Modular Monolingual Adaptation using Pretrained Language Models
Replacing tokens, freezing the corresponding embeddings, and tuning the rest of the model improves NLU performance on low-resource languages compared to full fine-tuning.