Lost in the Mix: Evaluating LLM Understanding of Code-Switched Text

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mohamed, Amr, Zhang, Yang, Vazirgiannis, Michalis, Shang, Guokan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909650370166784
author Mohamed, Amr
Zhang, Yang
Vazirgiannis, Michalis
Shang, Guokan
author_facet Mohamed, Amr
Zhang, Yang
Vazirgiannis, Michalis
Shang, Guokan
contents Code-switching (CSW) is the act of alternating between two or more languages within a single discourse. This phenomenon is widespread in multilingual communities, and increasingly prevalent in online content, where users naturally mix languages in everyday communication. As a result, Large Language Models (LLMs), now central to content processing and generation, are frequently exposed to code-switched inputs. Given their widespread use, it is crucial to understand how LLMs process and reason about such mixed-language text. This paper presents a systematic evaluation of LLM comprehension under code-switching by generating CSW variants of established reasoning and comprehension benchmarks. While degradation is evident when foreign tokens disrupt English text$\unicode{x2013}$even under linguistic constraints$\unicode{x2013}$embedding English into other languages often improves comprehension. Though prompting yields mixed results, fine-tuning offers a more stable path to degradation mitigation.
format Preprint
id arxiv_https___arxiv_org_abs_2506_14012
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Lost in the Mix: Evaluating LLM Understanding of Code-Switched Text
Mohamed, Amr
Zhang, Yang
Vazirgiannis, Michalis
Shang, Guokan
Computation and Language
Code-switching (CSW) is the act of alternating between two or more languages within a single discourse. This phenomenon is widespread in multilingual communities, and increasingly prevalent in online content, where users naturally mix languages in everyday communication. As a result, Large Language Models (LLMs), now central to content processing and generation, are frequently exposed to code-switched inputs. Given their widespread use, it is crucial to understand how LLMs process and reason about such mixed-language text. This paper presents a systematic evaluation of LLM comprehension under code-switching by generating CSW variants of established reasoning and comprehension benchmarks. While degradation is evident when foreign tokens disrupt English text$\unicode{x2013}$even under linguistic constraints$\unicode{x2013}$embedding English into other languages often improves comprehension. Though prompting yields mixed results, fine-tuning offers a more stable path to degradation mitigation.
title Lost in the Mix: Evaluating LLM Understanding of Code-Switched Text
topic Computation and Language
url https://arxiv.org/abs/2506.14012