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

arXiv:2601.06767v2 Announce Type: replace-cross Abstract: 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 +7 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.

Related

• AdaMCoT: Rethinking Cross-Lingual Factual Reasoning through Adaptive Multilingual Chain-of-Thought
• On the Step Length Confounding in LLM Reasoning Data Selection
• A Survey of Inductive Reasoning for Large Language Models
• Putting the Value Back in RL: Better Test-Time Scaling by Unifying LLM Reasoners With Verifiers
• SHAPE: Stage-aware Hierarchical Advantage via Potential Estimation for LLM Reasoning
• Rectifying LLM Thought from Lens of Optimization

Source: arXiv cs.AI | 2026-04-14