Cover Image

Making Qwen3 Think in Korean with Reinforcement Learning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Lee, Jungyup, Kim, Jemin, Park, Sang, Lee, SeungJae
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Smoothie-Qwen: Post-Hoc Smoothing to Reduce Language Bias in Multilingual LLMs
by: Ji, SeungWon, et al.
Published: (2025)

DNA 1.0 Technical Report
by: Lee, Jungyup, et al.
Published: (2025)

Handling Korean Out-of-Vocabulary Words with Phoneme Representation Learning
by: Kim, Nayeon, et al.
Published: (2025)

A Mean-Reverting Model of Exchange Rate Risk Premium Using Ornstein-Uhlenbeck Dynamics
by: Hwang, SeungJae
Published: (2025)

KOMBO: Korean Character Representations Based on the Combination Rules of Subcharacters
by: Kim, SungHo, et al.
Published: (2026)

SCRIPT: A Subcharacter Compositional Representation Injection Module for Korean Pre-Trained Language Models
by: Kim, SungHo, et al.
Published: (2026)

SDS KoPub VDR: A Benchmark Dataset for Visual Document Retrieval in Korean Public Documents
by: Lee, Jaehoon, et al.
Published: (2025)

A Survey on Inference Engines for Large Language Models: Perspectives on Optimization and Efficiency
by: Park, Sihyeong, et al.
Published: (2025)

Pragmatic Competence Evaluation of Large Language Models for the Korean Language
by: Park, Dojun, et al.
Published: (2024)

Thunder-DeID: Accurate and Efficient De-identification Framework for Korean Court Judgments
by: Hahm, Sungeun, et al.
Published: (2025)

Thunder-LLM: Efficiently Adapting LLMs to Korean with Minimal Resources
by: Kim, Jinpyo, et al.
Published: (2025)

Evaluating Large language models on Understanding Korean indirect Speech acts
by: Koo, Youngeun, et al.
Published: (2025)

Polishing Every Facet of the GEM: Testing Linguistic Competence of LLMs and Humans in Korean
by: Kim, SungHo, et al.
Published: (2025)

Do LLMs Need Inherent Reasoning Before Reinforcement Learning? A Study in Korean Self-Correction
by: Kim, Hongjin, et al.
Published: (2026)

KoCoNovel: Annotated Dataset of Character Coreference in Korean Novels
by: Kim, Kyuhee, et al.
Published: (2024)

Mentor-KD: Making Small Language Models Better Multi-step Reasoners
by: Lee, Hojae, et al.
Published: (2024)

K/DA: Automated Data Generation Pipeline for Detoxifying Implicitly Offensive Language in Korean
by: Jeon, Minkyeong, et al.
Published: (2025)

Qwen3 Technical Report
by: Yang, An, et al.
Published: (2025)

K-Act2Emo: Korean Commonsense Knowledge Graph for Indirect Emotional Expression
by: Kim, Kyuhee, et al.
Published: (2024)

KVoiceBench, KOpenAudioBench, and KMMAU: Agent-Driven Korean Speech Benchmarks for Evaluating SpeechLMs
by: Kim, Haechan, et al.
Published: (2026)

Open Ko-LLM Leaderboard2: Bridging Foundational and Practical Evaluation for Korean LLMs
by: Kim, Hyeonwoo, et al.
Published: (2024)

Can Code-Switched Texts Activate a Knowledge Switch in LLMs? A Case Study on English-Korean Code-Switching
by: Kim, Seoyeon, et al.
Published: (2024)

Open Ko-LLM Leaderboard: Evaluating Large Language Models in Korean with Ko-H5 Benchmark
by: Park, Chanjun, et al.
Published: (2024)

Think, Verbalize, then Speak: Bridging Complex Thoughts and Comprehensible Speech
by: Woo, Sang Hoon, et al.
Published: (2025)

Qwen3Guard Technical Report
by: Zhao, Haiquan, et al.
Published: (2025)

Exploring the Trade-Offs: Quantization Methods, Task Difficulty, and Model Size in Large Language Models From Edge to Giant
by: Lee, Jemin, et al.
Published: (2024)

KorNAT: LLM Alignment Benchmark for Korean Social Values and Common Knowledge
by: Lee, Jiyoung, et al.
Published: (2024)

K-BrowseComp: A Web Browsing Agent Benchmark Grounded in Korean Contexts
by: Lee, Nahyun, et al.
Published: (2026)

HAE-RAE Bench: Evaluation of Korean Knowledge in Language Models
by: Son, Guijin, et al.
Published: (2023)

KMMLU: Measuring Massive Multitask Language Understanding in Korean
by: Son, Guijin, et al.
Published: (2024)

QwenLong-L1: Towards Long-Context Large Reasoning Models with Reinforcement Learning
by: Wan, Fanqi, et al.
Published: (2025)

Linguistically Informed Graph Model and Semantic Contrastive Learning for Korean Short Text Classification
by: Yoo, JaeGeon, et al.
Published: (2026)

TranSentence: Speech-to-speech Translation via Language-agnostic Sentence-level Speech Encoding without Language-parallel Data
by: Kim, Seung-Bin, et al.
Published: (2024)

Qwen3-Coder-Next Technical Report
by: Cao, Ruisheng, et al.
Published: (2026)

Qwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and Ranking
by: Li, Mingxin, et al.
Published: (2026)

Rewarding How Models Think Pedagogically: Integrating Pedagogical Reasoning and Thinking Rewards for LLMs in Education
by: Lee, Unggi, et al.
Published: (2026)

Thinking with DistilQwen: A Tale of Four Distilled Reasoning and Reward Model Series
by: Cai, Wenrui, et al.
Published: (2025)

Optimizing Language Augmentation for Multilingual Large Language Models: A Case Study on Korean
by: Choi, ChangSu, et al.
Published: (2024)

Nunchi-Bench: Benchmarking Language Models on Cultural Reasoning with a Focus on Korean Superstition
by: Kim, Kyuhee, et al.
Published: (2025)

Thunder-KoNUBench: A Corpus-Aligned Benchmark for Korean Negation Understanding
by: Jung, Sungmok, et al.
Published: (2026)