dorsal/arxiv
View SchemaDo LLMs Need Inherent Reasoning Before Reinforcement Learning? A Study in Korean Self-Correction
| Authors | Hongjin Kim, Jaewook Lee, Kiyoung Lee, Jong-hun Shin, Soojong Lim, Oh-Woog Kwon |
|---|---|
| Categories | |
| ArXiv ID | 2601.05459vv1 |
| URL | https://arxiv.org/abs/2601.05459 |
| License | http://creativecommons.org/licenses/by/4.0/ |
Abstract
Large Language Models (LLMs) demonstrate strong reasoning and self-correction abilities in high-resource languages like English, but their performance remains limited in low-resource languages such as Korean. In this study, we investigate whether reinforcement learning (RL) can enhance Korean reasoning abilities to a degree comparable to English. Our findings reveal that RL alone yields limited improvements when applied to models lacking inherent Korean reasoning capabilities. To address this, we explore several fine-tuning strategies and show that aligning the model's internal reasoning processes with Korean inputs-particularly by tuning Korean-specific neurons in early layers-is key to unlocking RL's effectiveness. We introduce a self-correction code-switching dataset to facilitate this alignment and observe significant performance gains in both mathematical reasoning and self-correction tasks. Ultimately, we conclude that the crucial factor in multilingual reasoning enhancement is not injecting new linguistic knowledge, but effectively eliciting and aligning existing reasoning capabilities. Our study provides a new perspective on how internal translation and neuron-level tuning contribute to multilingual reasoning alignment in LLMs.
{
"annotation_id": "a6c6b37c-fdd2-4292-87b6-85718db44e00",
"date_created": "2026-02-17T05:53:03.790000Z",
"date_modified": "2026-02-17T05:53:03.790000Z",
"file_hash": "5861e7b48c59e76ecc353377486f9b1f30cb02f81a908a12526f9874d77f11b2",
"private": false,
"record": {
"abstract": "Large Language Models (LLMs) demonstrate strong reasoning and self-correction abilities in high-resource languages like English, but their performance remains limited in low-resource languages such as Korean. In this study, we investigate whether reinforcement learning (RL) can enhance Korean reasoning abilities to a degree comparable to English. Our findings reveal that RL alone yields limited improvements when applied to models lacking inherent Korean reasoning capabilities. To address this, we explore several fine-tuning strategies and show that aligning the model\u0027s internal reasoning processes with Korean inputs-particularly by tuning Korean-specific neurons in early layers-is key to unlocking RL\u0027s effectiveness. We introduce a self-correction code-switching dataset to facilitate this alignment and observe significant performance gains in both mathematical reasoning and self-correction tasks. Ultimately, we conclude that the crucial factor in multilingual reasoning enhancement is not injecting new linguistic knowledge, but effectively eliciting and aligning existing reasoning capabilities. Our study provides a new perspective on how internal translation and neuron-level tuning contribute to multilingual reasoning alignment in LLMs.",
"arxiv_id": "2601.05459",
"authors": [
"Hongjin Kim",
"Jaewook Lee",
"Kiyoung Lee",
"Jong-hun Shin",
"Soojong Lim",
"Oh-Woog Kwon"
],
"categories": [
"cs.CL",
"cs.AI"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"title": "Do LLMs Need Inherent Reasoning Before Reinforcement Learning? A Study in Korean Self-Correction",
"url": "https://arxiv.org/abs/2601.05459",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "f73c66f1-de4d-41e1-9660-ed0027199e96",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}