paper-with-me

Papers

MAPO: Advancing Multilingual Reasoning through Multilingual Alignment-as-Preference Optimization

2024-01-12 · Shuaijie She, Wei Zou, ShuJian Huang, Wenhao Zhu, Xiang Liu, Xiang Geng, Jiajun Chen

Though reasoning abilities are considered language-agnostic, existing LLMs exhibit inconsistent reasoning abilities across different languages, e.g., reasoning in the dominant language like English is superior to other languages due to the imbalance of multilingual training data. To enhance reasoning abilities in non-dominant languages, we propose a Multilingual-Alignment-as-Preference Optimization framework (MAPO), aiming to align the reasoning processes in other languages with the dominant language. Specifically, we harness an off-the-shelf translation model for the consistency between answers in non-dominant and dominant languages, which we adopt as the preference for optimization, e.g., Direct Preference Optimization (DPO) or Proximal Policy Optimization (PPO). Experiments show that MAPO stably achieves significant improvements in the multilingual reasoning of various models on all three benchmarks (MSVAMP +16.2%, MGSM +6.1%, and MNumGLUESub +13.3%), with improved reasoning consistency across languages.

📄 PDF Abstract BibTeX arXiv:2401.06838

Code (1)

njunlp/mapo 공식 구현 pytorch

Tasks

Mathematical Reasoning

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…
Entropy Regularization 설명 없음
PPO Proximal Policy Optimization, or PPO, is a policy gradient method for reinforcement learning. The motivation was to have an algorithm with the data efficiency and reliable…

Similar Papers 제목 키워드 기반

Cross-Lingual Consistency: A Novel Inference Framework for Advancing Reasoning in Large Language Models

2025-04-02 · Zhiwei Yu, Tuo Li, Changhong Wang, Hui Chen 외

Chain-of-thought (CoT) has emerged as a critical mechanism for enhancing reasoning capabilities in large language models (LLMs), with self-consistency demonstrating notable promise in boosting performance. However, inher…

Math

LinguaLIFT: An Effective Two-stage Instruction Tuning Framework for Low-Resource Language Tasks

2024-12-17 · Hongbin Zhang, Kehai Chen, Xuefeng Bai, Yang Xiang 외

Large language models (LLMs) have demonstrated impressive multilingual understanding and reasoning capabilities, driven by extensive pre-training multilingual corpora and fine-tuning instruction data. However, a performa…

Math

MMATH: A Multilingual Benchmark for Mathematical Reasoning

2025-05-25 · Wenyang Luo, Wayne Xin Zhao, Jing Sha, Shijin Wang 외

The advent of large reasoning models, such as OpenAI o1 and DeepSeek R1, has significantly advanced complex reasoning tasks. However, their capabilities in multilingual complex reasoning remain underexplored, with existi…

MathMathematical Reasoning

Seed-X: Building Strong Multilingual Translation LLM with 7B Parameters

2025-07-18 · Shanbo Cheng, Yu Bao, Qian Cao, Luyang Huang 외 arxiv

Multilingual translation stands as a challenging task for large language models (LLMs) to handle intricate language patterns and stilted translations that arise in automated translations. In this paper, we introduce Seed…

Reinforcement Learning

HistoryBankQA: Multilingual Temporal Question Answering on Historical Events

2025-09-16 · Biswadip Mandal, Anant Khandelwal, Manish Gupta arxiv

Temporal reasoning about historical events is a critical skill for NLP tasks like event extraction, historical entity linking, temporal question answering, timeline summarization, temporal event clustering and temporal n…

Natural Language UnderstandingNatural Language InferenceTimeline SummarizationQuestion Answering