Research Papers 论文研究 7h ago Updated 3h ago 更新于 3小时前 46

Benchmarking Language Models for Statistical Problem Formulation 语言模型统计问题表述基准测试

The paper formalizes "Statistical Problem Formulation" as a critical upstream step where LLMs must infer the implied statistical task and relevant variables from informal user goals and heterogeneous data StatFormBench is introduced as a new benchmark containing 1,013 samples across 20 coarse-grained and 85 fine-grained statistical problem categories, derived from five cross-domain statistics textbooks and a data science case library The best zero-shot LLMs achieve only 72.0% fine-grained classi 首次将LLM在统计工作中的"上游"问题表述环节形式化为独立任务,分解为统计问题分类和变量识别与角色分配两个子任务 发布StatFormBench基准测试,包含1013个样本,覆盖20个粗粒度和85个细粒度统计问题类别,源自五本跨领域统计教材和案例库 14个主流LLM的零样本评估显示,最佳模型细粒度分类准确率仅72.0%,变量集重叠仅63.2%,存在显著能力缺口 增强提示策略仅带来有限或不一致的改进,且无模型在两个子任务上均表现最优

60
Hot 热度
72
Quality 质量
65
Impact 影响力

Analysis 深度分析

TL;DR

  • The paper formalizes "Statistical Problem Formulation" as a critical upstream step where LLMs must infer the implied statistical task and relevant variables from informal user goals and heterogeneous data
  • StatFormBench is introduced as a new benchmark containing 1,013 samples across 20 coarse-grained and 85 fine-grained statistical problem categories, derived from five cross-domain statistics textbooks and a data science case library
  • The best zero-shot LLMs achieve only 72.0% fine-grained classification accuracy and 63.2% variable set overlap, revealing significant gaps in current model capabilities
  • No single model consistently outperforms others across both subtasks (Problem Classification and Variable Identification & Role Assignment), suggesting task-specific strengths
  • Enhanced prompting strategies yield only limited or inconsistent improvements, indicating fundamental challenges rather than easily solvable presentation issues

Why It Matters

This research addresses a critical gap in LLM evaluation: while many benchmarks assume the analysis target is pre-specified, real-world users approach models with vague goals and raw data, requiring the model to first formulate the statistical problem. For AI practitioners building data science assistants, these findings highlight that current models struggle with the foundational step of understanding what analysis is actually needed before any computation begins.

Technical Details

  • Statistical Problem Formulation is decomposed into two subtasks: (1) Statistical Problem Classification — identifying the type of statistical analysis implied by the user's description, and (2) Variable Identification & Role Assignment — determining which variables in the dataset are relevant and what roles they play (dependent, independent, control, etc.)
  • StatFormBench contains 1,013 samples spanning diverse problem types, data representations, and scenario styles, with a hierarchical taxonomy of 20 coarse-grained and 85 fine-grained statistical problem categories
  • The benchmark was constructed from five cross-domain statistics textbooks and a data science case library, ensuring broad coverage of real-world statistical scenarios
  • Evaluation covered 14 open- and closed-source LLMs, testing both zero-shot and enhanced prompting configurations
  • The benchmark data is released on Hugging Face and evaluation code on GitHub for reproducibility

Industry Insight

  • AI tool builders should prioritize improving the "problem formulation" stage in their pipelines, as this upstream reasoning step remains a significant bottleneck for end-to-end data science automation
  • The inconsistent performance across models on different subtasks suggests that specialized fine-tuning or modular architectures may outperform general-purpose prompting for statistical assistance applications
  • The limited gains from enhanced prompting indicate that fundamental architectural or training improvements are needed rather than superficial interface changes for this capability

TL;DR

  • 首次将LLM在统计工作中的"上游"问题表述环节形式化为独立任务,分解为统计问题分类和变量识别与角色分配两个子任务
  • 发布StatFormBench基准测试,包含1013个样本,覆盖20个粗粒度和85个细粒度统计问题类别,源自五本跨领域统计教材和案例库
  • 14个主流LLM的零样本评估显示,最佳模型细粒度分类准确率仅72.0%,变量集重叠仅63.2%,存在显著能力缺口
  • 增强提示策略仅带来有限或不一致的改进,且无模型在两个子任务上均表现最优

为什么值得看

该研究填补了LLM评估中"分析目标已明确"假设的盲区,揭示了模型在实际统计辅助场景中的关键瓶颈。对AI从业者而言,这为构建更实用的数据科学助手指明了技术改进方向。

技术解析

  • 任务定义:将"统计问题表述"(Statistical Problem Formulation)形式化为两个子任务——(1) 统计问题分类:从用户非正式目标推断隐含的统计任务;(2) 变量识别与角色分配:确定相关数据变量及其在统计模型中的角色
  • 基准构建:StatFormBench从五本跨领域统计教材和数据科学案例库构建,覆盖多样化问题类型、数据表示和场景风格,包含1013个样本,划分为20个粗粒度和85个细粒度类别
  • 实验设置:评估14个开源和闭源LLM的零样本表现,同时测试增强提示策略的效果
  • 核心指标:细粒度分类准确率(最佳72.0%)和变量集重叠度(最佳63.2%)

行业启示

  • LLM在统计/数据科学辅助领域的应用仍停留在"执行层",在"理解与定义问题"这一上游环节存在显著能力不足,需针对性改进
  • 提示工程并非万能解药,对于复杂的语义理解和任务分解任务,可能需要架构层面的创新(如专用模块、工具调用或微调)
  • 建议AI产品团队在构建数据科学助手时,将问题表述能力纳入核心评估指标,而非仅关注下游分析任务的执行质量

Disclaimer: The above content is generated by AI and is for reference only. 免责声明:以上内容由 AI 生成,仅供参考。

LLM 大模型 Benchmark 基准测试 Evaluation 评测 Research 科学研究