Ability Transfer Through Language Mixing

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61988987%3A17610%2F25%3AA2603C62" target="_blank" >RIV/61988987:17610/25:A2603C62 - isvavai.cz</a>

Result on the web

<a href="https://aclanthology.org/2025.ijcnlp-long.76/" target="_blank" >https://aclanthology.org/2025.ijcnlp-long.76/</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Ability Transfer Through Language Mixing

Original language description

We systematically investigate cross-lingual ability transfer in language models through controlled experiments across three problem sets: algorithmic addition, graph navigation, and natural language modeling. Our experimental design creates high-resource and low-resource "language" pairs differing in vocabulary, grammar, and computational requirements. We show that training on mixed datasets consistently enables strong positive transfer, significantly improving low-resource language performance compared to training on low amount of data in isolation. We observe improvements from 0% to 100% accuracy in arithmetic tasks, from 24% to 98% accuracy in graph navigation tasks, and 69.6% perplexity reduction in natural language modeling. We demonstrate that transfer effectiveness depends on computational complexity and linguistic differences, where grammar modifications support stronger transfer than vocabulary modifications. These findings provide compelling evidence that cross-lingual ability transfer is a robust mechanism which contributes to the quality of large language models in low-resource languages.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

<a href="/en/project/EH23_025%2F0008724" target="_blank" >EH23_025/0008724: Biography of Fake News with a Touch of AI: Dangerous Phenomenon through the Prism of Modern Human Sciences</a><br>

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Publication year

2025

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)

ISBN

979-8-89176-298-5

ISSN

—

e-ISSN

—

Number of pages

8

Pages from-to

1374-1381

Publisher name

Association for Computational Linguistics

Place of publication

—

Event location

Indie, Bombay

Event date

Dec 20, 2025

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—