CommonLID: Re-evaluating state-of-the-art language identification performance on web data
A community-driven, human-annotated LID benchmark for the web domain, covering 109 languages.
Topics:
Language identification (LID) is a fundamental step in curating multilingual corpora. However, LID models still perform poorly for many languages, especially on the noisy and heterogeneous web data often used to train multilingual language models. In this paper, we introduce CommonLID, a community-driven, human-annotated LID benchmark for the web domain, covering 109 languages. Many of the included languages have been previously under-served, making CommonLID a key resource for developing more representative high-quality text corpora. We show CommonLID's value by using it, alongside five other common evaluation sets, to test eight popular LID models. We analyse our results to situate our contribution and to provide an overview of the state of the art. In particular, we highlight that existing evaluations overestimate LID accuracy for many languages in the web domain. We make CommonLID and the code used to create it available under an open, permissive license.
Latest publications
Routing with generated data: Annotation-free LLM skill estimation and expert selection
A setting in which routers are trained on generated queries and answers produced from high-level task descriptions.
ACLMacaron: Controlled, human-written benchmark for multilingual and multicultural reasoning via template-filling
A template-first benchmark that factorizes reasoning type and cultural aspect across question languages.
ACLTemporal tokenization strategies for event sequence modeling with Large Language Models
A study of temporal tokenization for modeling event sequences with LLMs, comparing distinct encoding strategies.
ACL