paper-with-me

홈 › Papers

Efficiently Adapting Pretrained Language Models To New Languages

2023-11-09 · Zoltan Csaki, Pian Pawakapan, Urmish Thakker, Qiantong Xu

Recent large language models (LLM) exhibit sub-optimal performance on low-resource languages, as the training data of these models is usually dominated by English and other high-resource languages. Furthermore, it is challenging to train models for low-resource languages, especially from scratch, due to a lack of high quality training data. Adapting pretrained LLMs reduces the need for data in the new language while also providing cross lingual transfer capabilities. However, naively adapting to new languages leads to catastrophic forgetting and poor tokenizer efficiency. In this work, we study how to efficiently adapt any existing pretrained LLM to a new language without running into these issues. In particular, we improve the encoding efficiency of the tokenizer by adding new tokens from the target language and study the data mixing recipe to mitigate forgetting. Our experiments on adapting an English LLM to Hungarian and Thai show that our recipe can reach better performance than open source models on the target language, with minimal regressions on English.

📄 PDF Abstract BibTeX arXiv:2311.05741

Code (0)

등록된 구현이 없습니다.

Tasks

Cross-Lingual Transfer

Similar Papers 제목 키워드 기반

Adapting BigScience Multilingual Model to Unseen Languages

2022-04-11 · Zheng-Xin Yong, Vassilina Nikoulina

We benchmark different strategies of adding new languages (German and Korean) into the BigScience's pretrained multilingual language model with 1.3 billion parameters that currently supports 13 languages. We investigate …

Language ModelingLanguage Modellingmodel

Modular Monolingual Adaptation using Pretrained Language Models

2026-06-04 · Nalin Kumar, Ondřej Dušek arxiv

Building monolingual language models (LMs) for low-resource languages typically relies on adapting pretrained language models (PLMs) by finetuning the whole model on the target language. This approach is widely favored o…

Natural Language Understanding

Train Global, Tailor Local: Minimalist Multilingual Translation into Endangered Languages

2023-05-05 · Zhong Zhou, Jan Niehues, Alex Waibel

In many humanitarian scenarios, translation into severely low resource languages often does not require a universal translation engine, but a dedicated text-specific translation engine. For example, healthcare records, h…

HumanitarianTranslation

UNKs Everywhere: Adapting Multilingual Language Models to New Scripts

2020-12-31 · EMNLP 2021 11 · Jonas Pfeiffer, Ivan Vulić, Iryna Gurevych, Sebastian Ruder

Massively multilingual language models such as multilingual BERT offer state-of-the-art cross-lingual transfer performance on a range of NLP tasks. However, due to limited capacity and large differences in pretraining da…

Cross-Lingual Transfer

Exploring Pretraining via Active Forgetting for Improving Cross Lingual Transfer for Decoder Language Models

2024-10-21 · Divyanshu Aggarwal, Ashutosh Sathe, Sunayana Sitaram

Large Language Models (LLMs) demonstrate exceptional capabilities in a multitude of NLP tasks. However, the efficacy of such models to languages other than English is often limited. Prior works have shown that encoder-on…

Cross-Lingual TransferDecoder