This paper presents an extension of the Stochastic Answer Network (SAN), one
of the state-of-the-art machine reading comprehension models, to be able to
judge whether a question is unanswerable or not. The extended SAN contains two
components: a span detector and a binary classifier for judging whether the
question is unanswerable, and both components are jointly optimized.
Experiments show that SAN achieves the results competitive to the
state-of-the-art on Stanford Question Answering Dataset (SQuAD) 2.0. To
facilitate the research on this field, we release our code:
this https URL

该论文介绍了一种扩展的 Stochastic Answer Network 模型，添加了一个判断问题是否无法回答的二元分类器和句子定位模块，并通过实验证明其在 Stanford Question Answering Dataset (SQuAD) 2.0 数据集上具有与最先进模型相当的效果。

SQuAD 2.0 的随机答案网络

Stochastic Answer Networks for SQuAD 2.0

We propose a stochastic answer network (SAN) to explore multi-step inference
strategies in Natural Language Inference. Rather than directly predicting the
results given the inputs, the model maintains a state and iteratively refines
its predictions. Our experiments show that SAN achieves the state-of-the-art
results on three benchmarks: Stanford Natural Language Inference (SNLI)
dataset, MultiGenre Natural Language Inference (MultiNLI) dataset and Quora
Question Pairs dataset.

本文提出了一种随机答案网络（SAN）来研究自然语言推理中的多步推理策略，并通过实验表明 SAN 在三个基准测试中均取得了最先进的结果，包括斯坦福自然语言推理（SNLI）数据集，多样化自然语言推理（MultiNLI）数据集和 Quora 问题对数据集。

自然语言推理的随机回答网络

Stochastic Answer Networks for Natural Language Inference

We propose a simple yet robust stochastic answer network (SAN) that simulates
multi-step reasoning in machine reading comprehension. Compared to previous
work such as ReasoNet which used reinforcement learning to determine the number
of steps, the unique feature is the use of a kind of stochastic prediction
dropout on the answer module (final layer) of the neural network during the
training. We show that this simple trick improves robustness and achieves
results competitive to the state-of-the-art on the Stanford Question Answering
Dataset (SQuAD), the Adversarial SQuAD, and the Microsoft MAchine Reading
COmprehension Dataset (MS MARCO).

提出了一种简单而健壮的随机答案网络（SAN），通过在神经网络的最后一层中使用一种随机预测辍学的方式，模拟机器阅读理解中的多步推理，以此提高鲁棒性并在 Stanford 问题回答数据集（SQuAD）、对抗性 SQuAD 和 Microsoft 机器阅读理解数据集（MS MARCO）上获得了具有竞争力的结果。