The Impact of Model Scaling on Seen and Unseen Language Performance

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pokharel, Rhitabrat, Nezhad, Sina Bagheri, Agrawal, Ameeta, Singh, Suresh
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909453193838592
author Pokharel, Rhitabrat
Nezhad, Sina Bagheri
Agrawal, Ameeta
Singh, Suresh
author_facet Pokharel, Rhitabrat
Nezhad, Sina Bagheri
Agrawal, Ameeta
Singh, Suresh
contents The rapid advancement of Large Language Models (LLMs), particularly those trained on multilingual corpora, has intensified the need for a deeper understanding of their performance across a diverse range of languages and model sizes. Our research addresses this critical need by studying the performance and scaling behavior of multilingual LLMs in text classification and machine translation tasks across 204 languages. We systematically examine both seen and unseen languages across three model families of varying sizes in zero-shot and few-shot settings. Our findings show significant differences in scaling behavior between zero-shot and two-shot scenarios, with striking disparities in performance between seen and unseen languages. Model scale has little effect on zero-shot performance, which remains mostly flat. However, in two-shot settings, larger models show clear linear improvements in multilingual text classification. For translation tasks, however, only the instruction-tuned model showed clear benefits from scaling. Our analysis also suggests that overall resource levels, not just the proportions of pretraining languages, are better predictors of model performance, shedding light on what drives multilingual LLM effectiveness.
format Preprint
id arxiv_https___arxiv_org_abs_2501_05629
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The Impact of Model Scaling on Seen and Unseen Language Performance
Pokharel, Rhitabrat
Nezhad, Sina Bagheri
Agrawal, Ameeta
Singh, Suresh
Computation and Language
Artificial Intelligence
The rapid advancement of Large Language Models (LLMs), particularly those trained on multilingual corpora, has intensified the need for a deeper understanding of their performance across a diverse range of languages and model sizes. Our research addresses this critical need by studying the performance and scaling behavior of multilingual LLMs in text classification and machine translation tasks across 204 languages. We systematically examine both seen and unseen languages across three model families of varying sizes in zero-shot and few-shot settings. Our findings show significant differences in scaling behavior between zero-shot and two-shot scenarios, with striking disparities in performance between seen and unseen languages. Model scale has little effect on zero-shot performance, which remains mostly flat. However, in two-shot settings, larger models show clear linear improvements in multilingual text classification. For translation tasks, however, only the instruction-tuned model showed clear benefits from scaling. Our analysis also suggests that overall resource levels, not just the proportions of pretraining languages, are better predictors of model performance, shedding light on what drives multilingual LLM effectiveness.
title The Impact of Model Scaling on Seen and Unseen Language Performance
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2501.05629