AdaMame: A Training Recipe for Adaptive Multilingual Reasoning
While Large Reasoning Models (LRMs) show strong performance in English, they often fail to reason in the language of the query, a phenomenon known as language collapse. Existing RL-based fixes typically add a binary language fidelity reward to the accuracy objective, yet still incur trade-off in accuracy, mid-trace code-switching, and excessive token usage. In this work, we propose AdaMame, a two-stage training recipe for multilingual mathematical reasoning that addresses these limitations by adaptively aligning the reasoning language to the query language without compromising accuracy. The first SFT stage fine-tunes on naturally occurring reasoning traces across five languages to establish multilingual reasoning capability. In the subsequent RL stage, we introduce AdaMame-GRPO, an adaptation of Group Relative Policy Optimization (GRPO) in which a query-conditioned alignment factor grows progressively during training, guiding the model to first explore diverse reasoning languages before exploiting reasoning in the query language. Evaluated across two benchmarks, two LRMs, and 12 languages, AdaMame-GRPO achieves Pareto-optimal performance across reasoning accuracy, language fidelity, and token efficiency over all baselines, with the strongest gains on out-of-domain, lower-resource languages.
Code (0)
등록된 구현이 없습니다.
Tasks
Mathematical ReasoningSimilar Papers 제목 키워드 기반
Think in English, Answer in Korean: Efficient Adaptation of Multilingual Tool-Using Agents
We present LuckyStar 111B, a 111B-parameter hybrid reasoning model developed through a collaboration between Cohere and LG CNS for Korean-English enterprise agents under practical memory and serving constraints. The mode…
Reinforcement LearningMathematical ReasoningGRPO Beyond English: A Large-Scale Study of GRPO in Non-English and Multilingual Settings
Reinforcement Learning with Verifiable Rewards (RLVR), often optimized with Group Relative Policy Optimization (GRPO), has become a central recipe for improving the reasoning capabilities of pretrained language models bu…
Reinforcement LearningReasoning as an Adaptive Defense for Safety
Reasoning methods that adaptively allocate test-time compute have advanced LLM performance on easy to verify domains such as math and code. In this work, we study how to utilize this approach to train models that exhibit…
Reinforcement LearningLLaMAX2: Your Translation-Enhanced Model also Performs Well in Reasoning
General Large Language Models (LLMs) excel in reasoning, but those enhanced for translation struggle with reasoning tasks. To address this, we propose a novel translationenhanced recipe that begins with instruct models a…
AdaCoT: Rethinking Cross-Lingual Factual Reasoning through Adaptive Chain-of-Thought
Large language models have shown impressive multilingual capabilities through pretraining on diverse corpora. While these models show strong reasoning abilities, their performance varies significantly across languages du…
Machine Translation