Rethinking KenLM: Good and Bad Model Ensembles for Efficient Text Quality Filtering in Large Web Corpora
Year: 2025
Type: conference-paper
Abstract: With the increasing demand for substantial amounts of high-quality data to train large language models (LLMs), efficiently filtering large web corpora has become a critical challenge.For this purpose, KenLM, a lightweight n-grambased language model that operates on CPUs, is widely used.However, the traditional method of training KenLM utilizes only high-quality data and, consequently, does not explicitly learn the linguistic patterns of low-quality data.To address this issue, we propose an ensemble approach that leverages two contrasting KenLMs: (i) Good KenLM, trained on high-quality data; and (ii) Bad KenLM, trained on low-quality data.Experimental results demonstrate that our approach significantly reduces noisy content while preserving high-quality content compared to the traditional KenLM training method.This indicates that our method can be a practical solution with minimal computational overhead for resource-constrained environments.
Authors: Yungi Kim, Hyunsoo Ha, Sukyung Lee, Jihoo Kim, Seonghoon Yang, Chanjun Park
Institutions: UpStage (South Korea), Korea University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4412889588