基于树状思维的结构化推理用于孟加拉语数学应用题
Structured Reasoning with Tree-of-Thoughts for Bengali Math Word Problems
AI总结：
本文提出基于树状思维的结构化推理方法，用于提升孟加拉语数学应用题的解决效果，通过实验验证ToT在中等至大规模模型中的有效性。
摘要：
数学应用题（MWPs）是自然语言处理中最具挑战性的任务之一，因为它们需要语言理解和多步数值推理。尽管链式思维（CoT）提示方法已显示出潜力，但其线性结构往往传播错误，限制了整体效果。为了解决这一限制，我们系统研究了使用SOMADHAN数据集的树状思维（ToT）推理方法，用于孟加拉语MWPs。由于计算和令牌成本限制，我们评估了100个代表性问题的精心挑选集，涵盖多个大型语言模型（LLMs），包括GPT-OSS和LLaMA变体，在标准提示、CoT和ToT策略下。我们的结果表明，CoT将基线准确率从78%（标准提示）提高到83%平均，而ToT进一步提高了性能，最高达到5个百分点，使用GPT-OSS-120B时达到88%的准确率。这些改进表明，ToT在中等至大规模模型中特别有效，但在较小模型中可能优势较小。总体而言，我们的发现确立了ToT作为解决低资源语言如孟加拉语数学问题的稳健框架。更广泛地说，这项研究表明，与CoT相比，结构化推理方法如ToT可以提供更可靠和全球一致的结果，为多语言NLP中的更好推理策略铺平道路。
英文摘要：
Mathematical Word Problems (MWPs) are among the most challenging tasks in natural language processing because they require both linguistic understanding and multi-step numerical reasoning. While Chain-of-Thought (CoT) prompting has shown promise, its linear structure often propagates errors, limiting overall effectiveness. To address this limitation, we present the a systematic study of Tree-of-Thought (ToT) reasoning for Bengali MWPs using the SOMADHAN dataset. Owing to computational and token-cost constraints, we evaluate a curated set of 100 representative problems across multiple large language models (LLMs), including GPT-OSS and LLaMA variants, under standard prompting, CoT, and ToT strategies. Our results show that CoT improves baseline accuracy from 78% (standard prompting) to 83% on average, while ToT further increases performance by up to 5 percentage points, achieving 88% accuracy with GPT-OSS-120B. These improvements highlight that ToT is particularly effective in medium-to-large-scale models but may offer less advantage for smaller ones. Overall, our findings establish ToT as a robust framework for solving mathematical problems in low-resource languages such as Bengali. More broadly, this study shows that structured reasoning methods like ToT can provide more reliable and globally consistent outcomes than CoT, paving the way for better reasoning strategies in multilingual NLP.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
