DIVERS-Bench: Evaluating Language Identification Across Domain Shifts and Code-Switching

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ojo, Jessica, Kamel, Zina, Adelani, David Ifeoluwa
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916960745291776
author Ojo, Jessica
Kamel, Zina
Adelani, David Ifeoluwa
author_facet Ojo, Jessica
Kamel, Zina
Adelani, David Ifeoluwa
contents Language Identification (LID) is a core task in multilingual NLP, yet current systems often overfit to clean, monolingual data. This work introduces DIVERS-BENCH, a comprehensive evaluation of state-of-the-art LID models across diverse domains, including speech transcripts, web text, social media texts, children's stories, and code-switched text. Our findings reveal that while models achieve high accuracy on curated datasets, performance degrades sharply on noisy and informal inputs. We also introduce DIVERS-CS, a diverse code-switching benchmark dataset spanning 10 language pairs, and show that existing models struggle to detect multiple languages within the same sentence. These results highlight the need for more robust and inclusive LID systems in real-world settings.
format Preprint
id arxiv_https___arxiv_org_abs_2509_17768
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DIVERS-Bench: Evaluating Language Identification Across Domain Shifts and Code-Switching
Ojo, Jessica
Kamel, Zina
Adelani, David Ifeoluwa
Computation and Language
Artificial Intelligence
Machine Learning
Language Identification (LID) is a core task in multilingual NLP, yet current systems often overfit to clean, monolingual data. This work introduces DIVERS-BENCH, a comprehensive evaluation of state-of-the-art LID models across diverse domains, including speech transcripts, web text, social media texts, children's stories, and code-switched text. Our findings reveal that while models achieve high accuracy on curated datasets, performance degrades sharply on noisy and informal inputs. We also introduce DIVERS-CS, a diverse code-switching benchmark dataset spanning 10 language pairs, and show that existing models struggle to detect multiple languages within the same sentence. These results highlight the need for more robust and inclusive LID systems in real-world settings.
title DIVERS-Bench: Evaluating Language Identification Across Domain Shifts and Code-Switching
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2509.17768