Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
41 tokens/sec
GPT-4o
59 tokens/sec
Gemini 2.5 Pro Pro
41 tokens/sec
o3 Pro
7 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Dictionaries to the Rescue: Cross-Lingual Vocabulary Transfer for Low-Resource Languages Using Bilingual Dictionaries (2506.01535v1)

Published 2 Jun 2025 in cs.CL and cs.AI

Abstract: Cross-lingual vocabulary transfer plays a promising role in adapting pre-trained LLMs to new languages, including low-resource languages. Existing approaches that utilize monolingual or parallel corpora face challenges when applied to languages with limited resources. In this work, we propose a simple yet effective vocabulary transfer method that utilizes bilingual dictionaries, which are available for many languages, thanks to descriptive linguists. Our proposed method leverages a property of BPE tokenizers where removing a subword from the vocabulary causes a fallback to shorter subwords. The embeddings of target subwords are estimated iteratively by progressively removing them from the tokenizer. The experimental results show that our approach outperforms existing methods for low-resource languages, demonstrating the effectiveness of a dictionary-based approach for cross-lingual vocabulary transfer.

User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (7)
  1. Haruki Sakajo (2 papers)
  2. Yusuke Ide (30 papers)
  3. Justin Vasselli (7 papers)
  4. Yusuke Sakai (36 papers)
  5. Yingtao Tian (32 papers)
  6. Hidetaka Kamigaito (62 papers)
  7. Taro Watanabe (76 papers)
X Twitter Logo Streamline Icon: https://streamlinehq.com