NAPG：面向混合表格-文本问答的非自回归程序生成
NAPG: Non-Autoregressive Program Generation for Hybrid Tabular-Textual Question Answering
AI总结：
本文提出非自回归程序生成框架NAPG，独立生成完整程序元组，解决自回归解码的错误传播和暴露偏差，在混合表格-文本问答中显著提升准确率和速度，达到新最先进性能。
摘要：
混合表格-文本问答（QA）需要从异构信息中进行推理，推理类型主要分为数值推理和跨度提取。当前的数值推理方法以自回归方式解码程序序列，每个解码步骤生成一个操作符或操作数。然而，逐步解码存在暴露偏差问题，并且随着解码步骤的展开，由于错误传播，程序生成的准确性急剧下降。在本文中，我们提出了一种非自回归程序生成框架，该框架独立生成包含操作符和操作数的完整程序元组，能够解决错误传播问题，同时显著提升程序生成速度。在ConvFinQA和MultiHiertt数据集上的实验表明，我们的非自回归程序生成方法相比强基线FinQANet（+5.06 Exe Acc和+4.80 Prog Acc百分点）和MT2Net（+7.97 EM和+6.38 F1百分点）带来了显著改进，建立了新的最先进性能，同时程序生成速度更快（21倍）。最后，随着数值推理步骤数量的增加，我们方法的性能下降幅度显著小于基线方法。我们的代码将很快公开。
英文摘要：
Hybrid tabular-textual question answering (QA) requires reasoning from heterogeneous information, and the types of reasoning are mainly divided into numerical reasoning and span extraction. Current numerical reasoning methods autoregressively decode program sequences, and each decoding step produces either an operator or an operand. However, the step-by-step decoding suffers from exposure bias, and the accuracy of program generation drops sharply as the decoding steps unfold due to error propagation. In this paper, we propose a non-autoregressive program generation framework, which independently generates complete program tuples containing both operators and operands, can address the error propagation issue while significantly boosting the speed of program generation. Experiments on the ConvFinQA and MultiHiertt datasets show that our non-autoregressive program generation method can bring about substantial improvements over the strong FinQANet (+5.06 Exe Acc and +4.80 Prog Acc points) and MT2Net (+7.97 EM and +6.38 F1 points) baselines, establishing the new state-of-the-art performance, while being much faster (21x) in program generation. Finally, with increasing numbers of numerical reasoning steps the performance drop of our method is significantly smaller than that of the baselines. Our code will be publicly available soon.
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。
