对比结构化解码与普通生成，发现前者更准但非万能。
Evaluating Structured Decoding for Text-to-Table Generation: Evidence from Three Datasets
- 用表格结构引导生成，提升结果准确性
- 在数值对齐任务中准确率提升显著，但在密集文本中反而下降
- 适合需要精确对齐的场景，如体育赛事数据生成
我们对大型语言模型（LLMs）在文本到表格生成中的结构化解码进行了全面评估。尽管以往研究多聚焦于无约束生成，但生成过程中施加结构约束的影响仍不明确。我们在三个不同基准数据集（E2E、Rotowire、Livesum）上，使用最大320亿参数的开源LLM，系统比较了基于模式引导（结构化）解码与标准单次提示的方法，在资源受限环境下的表现。评估涵盖单元格、行和表级别的多种指标。结果显示，结构化解码显著提升了生成表格的有效性和对齐度，尤其在需要精确数值对齐的任务（如Rotowire）中表现突出；但在涉及密集文本信息（E2E）或长文本聚合（Livesum）的场景下性能反而下降。我们进一步分析了不同评估指标的适用性，并讨论了模型规模的影响。
原文摘要 · Abstract (English)
We present a comprehensive evaluation of structured decoding for text-to-table generation with large language models (LLMs). While previous work has primarily focused on unconstrained generation of tables, the impact of enforcing structural constraints during generation remains underexplored. We systematically compare schema-guided (structured) decoding to standard one-shot prompting across three diverse benchmarks - E2E, Rotowire, and Livesum - using open-source LLMs of up to 32B parameters, assessing the performance of table generation approaches in resource-constrained settings. Our experiments cover a wide range of evaluation metrics at cell, row, and table levels. Results demonstrate that structured decoding significantly enhances the validity and alignment of generated tables, particularly in scenarios demanding precise numerical alignment (Rotowire), but may degrade performance in contexts involving densely packed textual information (E2E) or extensive aggregation over lengthy texts (Livesum). We further analyze the suitability of different evaluation metrics and discuss the influence of model size.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。