CORTEXA
← Browse
arxivcs.CLcs.AIcs.LG2026-07-11

Cost of Reasoning in non-English Languages: A Case Study on Japanese

Yuu Jinnai

Reasoning Language Models (RLMs) achieve their strongest performance when they reason in English, the language for which reasoning-oriented training data is most abundant. However, reasoning trace is a clue for model interpretability and safety, and useful in practice for both the model users and for model developers. Thus, it is desirable to be able to develop a model that reasons in a language of the user's choice, while still maintaining strong reasoning performance. To this end, we study the feasibility of training a model that reasons in Japanese. We develop a Japanese-reasoning variant of Qwen-3-Swallow-8B, which is a Japanese LLM continually pretrained from Qwen-3-8B, with GRPO and evaluate it across coding, math, and science benchmarks. The study shows that reasoning-language control is feasible by training a Japanese continually pretrained model with GRPO. However, its performance is at best on par with strong English-reasoning baselines on several benchmarks. We also evaluate the trained model on Japanese cultural benchmarks and observe that the model's performance is worse than the baseline models, suggesting that the reasoning in Japanese does not immediately improve performance on culturally relevant tasks for free.

View free PDFSource page

Related papers

arxivcs.AIcs.CLcs.ETcs.LGcs.MA2026-07-23

The Boundaries of Automation: A Theory of Persistent Human Participation

Fares Fourati, Hinrich Schütze, Eyke Hüllermeier, Iryna Gurevych

The rapid progress of AI has intensified the long-standing pursuit of automation: replacing human participation with algorithms wherever possible. Implicit in this pursuit is the assumption that humans remain in the loop only because current AI systems are not yet sufficiently ca…

View free PDFSource page
arxivcs.DBcs.AIcs.CLcs.LG2026-07-24

DBA-Bench: A Production-Fidelity Benchmark for LLM-Based Database Operations Agents

Junming Chen, Junyang Jiang, Xu Chen, Zibo Liang, Kai Zheng

LLM-based database agents show promise, but differing task scopes, testbeds, and metrics hinder comparison. We identify four gaps between evaluation and production operations: live-environment fidelity (multi-turn read-write interaction with a running database); observation-space…

View free PDFSource page
arxivcs.LGcs.AIcs.CL2026-07-24

Cross-Tokenizer On-Policy Distillation via Byte-Prefix Marginalization

Hao Wang, Kun Yuan, Wenlin Zhong, Minglei Zhang, Han Xiao, Ming Sun, et al.

Open-weight language models from different families exhibit complementary capabilities, motivating their consolidation into a compact student through on-policy distillation (OPD). However, full-vocabulary OPD typically assumes a shared tokenizer, while existing cross-tokenizer me…

View free PDFSource page
arxivcs.CLcs.AIcs.LG2026-07-31

CalibratedRubric: Task-Adaptive Rubric Banks for Open-Ended LLM Evaluation

Mengting Chen, Yanshu Sun, Wanting Liang, Beidi Luan, Rui Sun, Dezhi Chen, et al.

Reliable evaluation of open-ended LLM outputs requires fine-grained rubrics, yet expert curation is costly and difficult to scale. Existing automated pipelines rely on strict judge unanimity and binary variance filters, which cannot distinguish measurable rubrics from informative…

View free PDFSource page
arxivcs.LGcs.AIcs.CL2026-07-24

Teaching LLMs to Self-Evolve: Cultivating Core Meta-Skills with Reinforcement Learning

Shujin Wu, Cheng Qian, Xiusi Chen, Heng Ji

Test-time scaling through iterative self-evolution with environment feedback, as demonstrated by AlphaEvolve, shows remarkable performance gains. We hypothesize that the success of such evolution frameworks hinges on meta-skills, such as self-reflection with environment feedback,…

View free PDFSource page