mlsanityData to 7 Oct 2026

TAPO: Translation Augmented Policy Optimization for Multilingual Mathematical Reasoning

arXiv 2603.25419 ↗First posted 26 Mar 2026cs.CLcited by 6

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

Large Language Models (LLMs) have demonstrated remarkable proficiency in English mathematical reasoning, yet a significant performance disparity persists in multilingual contexts, largely attributed to deficiencies in language understanding. To bridge this gap, we introduce Translation-Augmented Policy Optimization (TAPO), a novel reinforcement learning framework built upon GRPO. TAPO enforces an explicit alignment strategy where the model leverages English as a pivot and follows an understand-then-reason paradigm. Crucially, we employ a step-level relative advantage mechanism that decouples understanding from reasoning, allowing the integration of translation quality rewards without introducing optimization conflicts. Extensive experiments reveal that TAPO effectively synergizes language understanding with reasoning capabilities and is compatible with various models. It outperforms baseline methods in both multilingual mathematical reasoning and translation tasks, while generalizing well to unseen languages and out-of-domain tasks.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 27 of its 50 references that are in the map. First posted on arXiv 26 Mar 2026.

Builds on

27 of 50 references.

Closest ancestors · 10

Same direction, newest first

  1. 2025Why Do Multilingual Reasoning Gaps Emerge in Reasoning Language Models?31 Oct
  2. 2025Think Natively: Unlocking Multilingual Reasoning with Consistency-Enhanced Reinforcement Learning8 Oct
  3. 2025Beyond English-Centric Training: How Reinforcement Learning Improves Cross-Lingual Reasoning in LLMs28 Sep
  4. 2025Learn Globally, Speak Locally: Bridging the Gaps in Multilingual Reasoning7 Jul
  5. 2025Cross-lingual Collapse: How Language-Centric Foundation Models Shape Reasoning in Large Language Models6 Jun
  6. 2025Language Matters: How Do Multilingual Input and Reasoning Paths Affect Large Reasoning Models?23 May
  7. 2025When Less Language is More: Language-Reasoning Disentanglement Makes LLMs Better Multilingual Reasoners21 May
  8. 2025Qwen3 Technical Report14 May
  9. 2025MT-R1-Zero: Advancing LLM-based Machine Translation via R1-Zero-like Reinforcement Learning14 Apr
  10. 2024DeepSeek-V3 Technical Report27 Dec

Other references · 17

  1. 2024The Llama 3 Herd of Modelscited by 14,192
  2. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  4. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  5. 2024GPT-4o System Cardcited by 4,467
  6. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
  7. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,678
  8. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 1,980
  9. 2024OpenAI o1 System Cardcited by 1,220
  10. 2025DeepSeek-V3.2: Pushing the Frontier of Open Large Language Modelscited by 755
  11. 2023RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignmentcited by 672
  12. 2025Group-in-Group Policy Optimization for LLM Agent Trainingcited by 442
5 more
  1. 2025RL's Razor: Why Online Reinforcement Learning Forgets Lesscited by 162
  2. 2025RLVMR: Reinforcement Learning with Verifiable Meta-Reasoning Rewards for Robust Long-Horizon Agentscited by 56
  3. 2023Language Versatilists vs. Specialists: An Empirical Revisiting on Multilingual Transfer Abilitycited by 27
  4. 2025MMATH: A Multilingual Benchmark for Mathematical Reasoningcited by 16
  5. 2025Making Qwen3 Think in Korean with Reinforcement Learningcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 6.

6 papers.

  1. 2026What Makes Good Multilingual Reasoning? Disentangling Traces with Measurable Featurescited by 4
  2. 2026Learning When to Translate for Multilingual Reasoningcited by 1
  3. 2026Translate-R1: Cost-Aware Translation Tool Use via Reinforcement Learningcited by 0
  4. 2026Soft Token Alignment for Cross-Lingual Reasoningcited by 0
  5. 2026RP-OPSD: Reasoning-Pivot-Guided On-Policy Self-Distillation for Multilingual Reasoning Transfercited by 0
  6. 2026GRPO Beyond English: A Large-Scale Study of GRPO in Non-English and Multilingual Settingscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026LANG: Reinforcement Learning for Multilingual Reasoning with Language-Adaptive Hint Guidance
  2. 2025Tool-Augmented Policy Optimization: Synergizing Reasoning and Adaptive Tool Use with Reinforcement Learning
  3. 2025Aligning Multilingual Reasoning with Verifiable Semantics from a High-Resource Expert Model
  4. 2026Cross-lingual Self-Consistency for Multilingual Reasoning with Language Models
  5. 2024MAPO: Advancing Multilingual Reasoning through Multilingual Alignment-as-Preference Optimization
  6. 2025Parallel Scaling Law: Unveiling Reasoning Generalization through A Cross-Linguistic Perspective
  7. 2026Efficient Multilingual Reasoning Transfer via Progressive Code-Switching
  8. 2026Transformation-Augmented GRPO for Enhancing Exploration in Reasoning of Large Language Models