# GanitLLM: Difficulty-Aware Bengali Mathematical Reasoning through Curriculum-GRPO

arXiv: https://arxiv.org/abs/2601.06767 · HTML: https://mlsanity.com/paper/2601.06767/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in Mathematical reasoning, one of 17 directions in LLM reasoning.
- Builds on 31 of its 51 references that are in the map.
- First posted on arXiv 11 Jan 2026.

## Placed in

- LLM reasoning / [Mathematical reasoning](https://mlsanity.com/lr/mathematical-reasoning/)

## Builds on (most cited first, 30 of 31)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025)
- [s1: Simple test-time scaling](https://mlsanity.com/paper/2501.19393/) (2025)
- [Phi-4 Technical Report](https://mlsanity.com/paper/2412.08905/) (2024)
- [The Lessons of Developing Process Reward Models in Mathematical Reasoning](https://mlsanity.com/paper/2501.07301/) (2025)
- [Competitive Programming with Large Reasoning Models](https://mlsanity.com/paper/2502.06807/) (2025)
- [Self-Evolving Curriculum for LLM Reasoning](https://mlsanity.com/paper/2505.14970/) (2025)
- [Breaking Language Barriers in Multilingual Mathematical Reasoning: Insights and Observations](https://mlsanity.com/paper/2310.20246/) (2023)
- [PolyMath: Evaluating Mathematical Reasoning in Multilingual Contexts](https://mlsanity.com/paper/2504.18428/) (2025)
- [ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline](https://mlsanity.com/paper/2404.02893/) (2024)
- [Prompt Curriculum Learning for Efficient LLM Post-Training](https://mlsanity.com/paper/2510.01135/) (2025)
- [Linguistic Generalizability of Test-Time Scaling in Mathematical Reasoning](https://mlsanity.com/paper/2502.17407/) (2025)
- [mCoT: Multilingual Instruction Tuning for Reasoning Consistency in Language Models](https://mlsanity.com/paper/2406.02301/) (2024)
- [Train Long, Think Short: Curriculum Learning for Efficient Reasoning](https://mlsanity.com/paper/2508.08940/) (2025)
- [TigerLLM - A Family of Bangla Large Language Models](https://mlsanity.com/paper/2503.10995/) (2025)
- [Transcending Language Boundaries: Harnessing LLMs for Low-Resource Language Translation](https://mlsanity.com/paper/2411.11295/) (2024)
- [UloRL:An Ultra-Long Output Reinforcement Learning Approach for Advancing Large Language Models' Reasoning Abilities](https://mlsanity.com/paper/2507.19766/) (2025)
- [TituLLMs: A Family of Bangla LLMs with Comprehensive Benchmarking](https://mlsanity.com/paper/2502.11187/) (2025)
- [Script Fragmentation and Format: What Drives the English-Bengali Performance Gap in Open LLMs?](https://mlsanity.com/paper/2507.23248/) (2025)
- [Leveraging Large Language Models for Bengali Math Word Problem Solving with Chain of Thought Reasoning](https://mlsanity.com/paper/2505.21354/) (2025)
- [From Data to Commonsense Reasoning: The Use of Large Language Models for Explainable AI](https://mlsanity.com/paper/2407.03778/) (2024)
- [Predictive Scaling Laws for Efficient GRPO Training of Large Reasoning Models](https://mlsanity.com/paper/2507.18014/) (2025)
- [Confucius3-Math: A Lightweight High-Performance Reasoning LLM for Chinese K-12 Mathematics Learning](https://mlsanity.com/paper/2506.18330/) (2025)
- [A Study on Thinking Patterns of Large Reasoning Models in Code Generation](https://mlsanity.com/paper/2509.13758/) (2025)
- [MojoBench: Language Modeling and Benchmarks for Mojo](https://mlsanity.com/paper/2410.17736/) (2024)

## Built on by (most cited first, 6 of 6)

- [Bridging Reasoning Trajectories in On-Policy Distillation via Near-Future Guidance](https://mlsanity.com/paper/2606.00305/) (2026)
- [Cornerstones or Stumbling Blocks? Deciphering the Rock Tokens in On-Policy Distillation](https://mlsanity.com/paper/2605.09253/) (2026)
- [PA3: Policy-Aware Agent Alignment through Chain-of-Thought](https://mlsanity.com/paper/2603.14602/) (2026)
- [Many Dialects, Many Languages, One Cultural Lens: Evaluating Multilingual VLMs for Bengali Culture Understanding Across Historically Linked Languages and Regional Dialects](https://mlsanity.com/paper/2603.21165/) (2026)
- [DecomposeRL: Learning to Ask Useful, Informative, and Diverse Questions for Semi-Supervised, Traceable Claim Verification](https://mlsanity.com/paper/2605.27858/) (2026)
- [Register Shifts Break LLM Safety: A Bengali Benchmark with Culturally Grounded Harms](https://mlsanity.com/paper/2608.22335/) (2026)

## Abstract (arXiv)

We present a Bengali mathematical reasoning model called GanitLLM (named after the Bangla word for mathematics, Ganit), together with a new difficulty-aware Bengali math corpus and a curriculum-based GRPO pipeline. Bengali is one of the world's most widely spoken languages, yet existing LLMs either reason in English and then translate, or simply fail on multi-step Bengali math, in part because reinforcement learning recipes are tuned for high-resource languages and collapse under reward sparsity in low-resource settings. To address this, we construct Ganit, a rigorously filtered and decontaminated Bengali math dataset with automatic difficulty tags derived from the pass@k of a strong evaluator model. Building on this dataset, we propose Curriculum-GRPO, which combines multi-stage training (SFT + GRPO) with difficulty-aware sampling and verifiable rewards for format, numerical correctness, and Bengali reasoning. On Bn-MGSM and Bn-MSVAMP, GanitLLM-4B improves over its Qwen3-4B base by +8 and +6 accuracy points, respectively, while increasing the percentage of Bengali reasoning tokens from 14% to over 88% and reducing average solution length from 943 to 193 words. Project page is available at https://dipta007.github.io/GanitLLM