IRIS: Interleaved Reinforcement with Incremental Staged Curriculum for Cross-Lingual Mathematical Reasoning

• URL: http://arxiv.org/abs/2604.24114v1
• Date: Mon, 27 Apr 2026 07:10:35 GMT
• Title: IRIS: Interleaved Reinforcement with Incremental Staged Curriculum for Cross-Lingual Mathematical Reasoning
• Abstract summary: IRIS is a two-axis framework that combinesSupervised Fine-Tuning on progressively harder problems with Reverse Curriculum Reinforcement Learning.<n>We release CL-Math, a dataset of 29k problems with step-level annotations in English, Hindi, and Marathi.<n>IRIS consistently improves performance across standard benchmarks and curated multilingual test sets.
• Score: 54.28485292978118
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Curriculum learning helps language models tackle complex reasoning by gradually increasing task difficulty. However, it often fails to generate consistent step-by-step reasoning, especially in multilingual and low-resource settings where cross-lingual transfer from English to Indian languages remains limited. We propose IRIS: Interleaved Reinforcement with Incremental Staged Curriculum, a two-axis framework that combines Supervised Fine-Tuning on progressively harder problems (vertical axis) with Reverse Curriculum Reinforcement Learning to reduce reliance on step-by-step guidance (horizontal axis). We design a composite reward combining correctness, step-wise alignment, continuity, and numeric incentives, optimized via Group Relative Policy Optimization (GRPO). We release CL-Math, a dataset of 29k problems with step-level annotations in English, Hindi, and Marathi. Across standard benchmarks and curated multilingual test sets, IRIS consistently improves performance, with strong results on math reasoning tasks and substantial gains in low-resource and bilingual settings, alongside modest improvements in high-resource languages.

Related papers

• Crosslingual On-Policy Self-Distillation for Multilingual Reasoning [48.68444770923683]
  Crosslingual On-Policy Self-Distillation (COPSD) transfers a model's own high-resource reasoning behavior to low-resource languages.<n>Experiments on 17 low-resource African languages show that COPSD consistently improves low-resource mathematical reasoning across model sizes.
  arXiv  Detail & Related papers  (2026-05-10T14:06:09Z)
• English is Not All You Need: Systematically Exploring the Role of Multilinguality in LLM Post-Training [7.141617235075284]
  Post-training pipelines remain predominantly English-centric, contributing to performance disparities across languages.<n>We present a systematic, controlled study of the interplay between training language coverage, model scale, and task domain.<n>We find that increasing language coverage during post-training is largely beneficial across tasks and model scales.
  arXiv  Detail & Related papers  (2026-04-14T20:26:34Z)
• CLEAR: Cross-Lingual Enhancement in Alignment via Reverse-training [25.752819016822574]
  Cross-Lingual Enhancement in Retrieval via Reverse-training (CLEAR) is a novel loss function utilizing a reverse training scheme to improve retrieval performance.<n>CLEAR achieves notable improvements in cross-lingual scenarios, with gains up to 15%, particularly in low-resource languages.
  arXiv  Detail & Related papers  (2026-04-07T12:54:38Z)
• Code-Switching In-Context Learning for Cross-Lingual Transfer of Large Language Models [64.54005959758733]
  We introduce code-switching in-context learning (CSICL) as a principled and robust approach for overcoming the translation barrier during inference.<n>We conduct extensive experiments across 4 LLMs, 6 datasets, and 10 languages, spanning both knowledge-intensive and reasoning-oriented domains.<n>Our results demonstrate CSICL consistently outperforms X-ICL baselines, achieving gains of 3.1%p and 1.9%p in both target and unseen languages.
  arXiv  Detail & Related papers  (2025-10-07T08:35:42Z)
• Improving Generative Cross-lingual Aspect-Based Sentiment Analysis with Constrained Decoding [0.8602553195689511]
  This paper introduces a novel approach using constrained decoding with sequence-to-sequence models.<n>It improves cross-lingual performance by 5% on average for the most complex task.<n>We evaluate our approach across seven languages and six ABSA tasks.
  arXiv  Detail & Related papers  (2025-08-14T06:07:53Z)
• Demystifying Multilingual Chain-of-Thought in Process Reward Modeling [86.98098988779809]
  We tackle the challenge of extending process reward models (PRMs) to multilingual settings.<n>We train multilingual PRMs on a dataset spanning seven languages, which is translated from English.<n>Our results highlight the sensitivity of multilingual PRMs to both the number of training languages and the volume of English data.
  arXiv  Detail & Related papers  (2025-02-18T09:11:44Z)
• No Train but Gain: Language Arithmetic for training-free Language Adapters enhancement [59.37775534633868]
  We introduce a novel method called language arithmetic, which enables training-free post-processing. The effectiveness of the proposed solution is demonstrated on three downstream tasks in a MAD-X-based set of cross-lingual schemes.
  arXiv  Detail & Related papers  (2024-04-24T08:52:40Z)
• Promoting Generalized Cross-lingual Question Answering in Few-resource Scenarios via Self-knowledge Distillation [2.2493846458264386]
  We study cross-lingual transfer mainly focusing on the Generalized Cross-Lingual Transfer (G-XLT) task. Our approach seeks to enhance cross-lingual QA transfer using a high-performing multilingual model trained on a large-scale dataset. We introduce the novel mAP@k coefficients to fine-tune self-knowledge distillation loss.
  arXiv  Detail & Related papers  (2023-09-29T10:54:59Z)
• Efficiently Aligned Cross-Lingual Transfer Learning for Conversational Tasks using Prompt-Tuning [98.60739735409243]
  Cross-lingual transfer of language models trained on high-resource languages like English has been widely studied for many NLP tasks. We introduce XSGD for cross-lingual alignment pretraining, a parallel and large-scale multilingual conversation dataset. To facilitate aligned cross-lingual representations, we develop an efficient prompt-tuning-based method for learning alignment prompts.
  arXiv  Detail & Related papers  (2023-04-03T18:46:01Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.