Targeted syntactic evaluations of language models ask whether models show stable preferences for syntactically acceptable content over minimal-pair unacceptable inputs. Most targeted syntactic evaluation datasets ask models to make these judgements with just a single context-free sentence as input. This does not match language models' training regime, in which input sentences are always highly contextualized by the surrounding corpus. This mismatch raises an important question: how robust are models' syntactic judgements in different contexts? In this paper, we investigate the stability of language models' performance on targeted syntactic evaluations as we vary properties of the input context: the length of the context, the types of syntactic phenomena it contains, and whether or not there are violations of grammaticality. We find that model judgements are generally robust when placed in randomly sampled linguistic contexts. However, they are substantially unstable for contexts containing syntactic structures matching those in the critical test content. Among all tested models (GPT-2 and five variants of OPT), we significantly improve models' judgements by providing contexts with matching syntactic structures, and conversely significantly worsen them using unacceptable contexts with matching but violated syntactic structures. This effect is amplified by the length of the context, except for unrelated inputs. We show that these changes in model performance are not explainable by simple features matching the context and the test inputs, such as lexical overlap and dependency overlap. This sensitivity to highly specific syntactic features of the context can only be explained by the models' implicit in-context learning abilities.

将已知的原始概念重组为更大的新型组合是一种典型的人类认知能力。NLP中的大型神经模型是否在从数据中学习时获得此能力是一个悬而未决的问题。在本文中，我们从形式语言的角度看一下这个问题。我们使用确定性有限状态传感器来制作具有控制组合性的可控属性的无限数量数据集。通过对许多传感器进行随机采样，我们探讨了它们的哪些属性（状态数，字母大小，过渡次数等）有助于通过神经网络的组成关系可学习。通常，我们发现模型要么完全学习关系。关键是过渡覆盖范围，以每个过渡为400个示例设置软可学习性限制。

机器学习（ML）研究通常集中在模型上，而最突出的数据集已用于日常的ML任务，而不考虑这些数据集对基本问题的广度，困难和忠诚。忽略数据集的基本重要性已引起了重大问题，该问题涉及现实世界中的数据级联以及数据集驱动标准的模型质量饱和，并阻碍了研究的增长。为了解决此问题，我们提出Dataperf，这是用于评估ML数据集和数据集工作算法的基准软件包。我们打算启用“数据棘轮”，其中培训集将有助于评估相同问题的测试集，反之亦然。这种反馈驱动的策略将产生一个良性的循环，该循环将加速以数据为中心的AI。MLCommons协会将维护Dataperf。

道德是人类最长的智力努力之一。近年来，AI和NLP的领域试图撰写与学习系统的与人类相互作用的学习系统，应该被限制为行为道德。该静脉中的一个提议是建立道德模型，可以采取任意文本，并输出关于所描述的情况的道德判断。在这项工作中，我们专注于对最近提出的Delphi模型的单一案例研究，并为该项目的建议自动化道德判决提供了批评。通过对Delphi的审计，我们检查更广泛的问题，适用于任何类似的尝试。我们讨论了机器道德如何通过专注于技术的当前和近期使用技术的方式来讨论机器伦理，以透明度，民主价值观，并允许直接的责任。

为了创建在广泛的测试输入中强大的模型，培训数据集应包括跨越许多现象的各种示例。动态的对抗数据收集（DADC），注释者制作的示例挑战不断改进模型的示例，其有望是生成这种多样化训练集的一种方法。先前的工作表明，在1-3轮中运行DADC可以帮助模型修复某些错误类型，但不一定会带来更好的概括，而不是对抗性测试数据。我们认为，在许多回合中运行DADC可以最大化其训练时间的好处，因为不同的回合可以涵盖许多与任务相关的现象。我们介绍了长期DADC的第一个研究，其中我们收集了20轮NLI示例，用于一小部分前提，并采用对抗性和非对抗性方法。与接受非对抗数据的训练的模型相比，接受DADC示例培训的模型在我们的专家策划测试集中的错误少26％。我们的分析表明，DADC产生的例子更加困难，更词法和句法多样性，并且与非对抗性示例相比，注释伪像更少。

We introduce a new large-scale NLI benchmark dataset, collected via an iterative, adversarial human-and-model-in-the-loop procedure. We show that training models on this new dataset leads to state-of-the-art performance on a variety of popular NLI benchmarks, while posing a more difficult challenge with its new test set. Our analysis sheds light on the shortcomings of current state-of-theart models, and shows that non-expert annotators are successful at finding their weaknesses. The data collection method can be applied in a never-ending learning scenario, becoming a moving target for NLU, rather than a static benchmark that will quickly saturate.

State-of-the-art natural language processing systems rely on supervision in the form of annotated data to learn competent models. These models are generally trained on data in a single language (usually English), and cannot be directly used beyond that language. Since collecting data in every language is not realistic, there has been a growing interest in crosslingual language understanding (XLU) and low-resource cross-language transfer. In this work, we construct an evaluation set for XLU by extending the development and test sets of the Multi-Genre Natural Language Inference Corpus (MultiNLI) to 15 languages, including low-resource languages such as Swahili and Urdu. We hope that our dataset, dubbed XNLI, will catalyze research in cross-lingual sentence understanding by providing an informative standard evaluation task. In addition, we provide several baselines for multilingual sentence understanding, including two based on machine translation systems, and two that use parallel data to train aligned multilingual bag-of-words and LSTM encoders. We find that XNLI represents a practical and challenging evaluation suite, and that directly translating the test data yields the best performance among available baselines.

This paper introduces the Multi-Genre Natural Language Inference (MultiNLI) corpus, a dataset designed for use in the development and evaluation of machine learning models for sentence understanding. At 433k examples, this resource is one of the largest corpora available for natural language inference (a.k.a. recognizing textual entailment), improving upon available resources in both its coverage and difficulty. MultiNLI accomplishes this by offering data from ten distinct genres of written and spoken English, making it possible to evaluate systems on nearly the full complexity of the language, while supplying an explicit setting for evaluating cross-genre domain adaptation. In addition, an evaluation using existing machine learning models designed for the Stanford NLI corpus shows that it represents a substantially more difficult task than does that corpus, despite the two showing similar levels of inter-annotator agreement. This task will involve reading a line from a non-fiction article and writing three sentences that relate to it. The line will describe a situation or event. Using only this description and what you know about the world:

Recent advances in upper limb prostheses have led to significant improvements in the number of movements provided by the robotic limb. However, the method for controlling multiple degrees of freedom via user-generated signals remains challenging. To address this issue, various machine learning controllers have been developed to better predict movement intent. As these controllers become more intelligent and take on more autonomy in the system, the traditional approach of representing the human-machine interface as a human controlling a tool becomes limiting. One possible approach to improve the understanding of these interfaces is to model them as collaborative, multi-agent systems through the lens of joint action. The field of joint action has been commonly applied to two human partners who are trying to work jointly together to achieve a task, such as singing or moving a table together, by effecting coordinated change in their shared environment. In this work, we compare different prosthesis controllers (proportional electromyography with sequential switching, pattern recognition, and adaptive switching) in terms of how they present the hallmarks of joint action. The results of the comparison lead to a new perspective for understanding how existing myoelectric systems relate to each other, along with recommendations for how to improve these systems by increasing the collaborative communication between each partner.

Heart failure remains a major public health challenge with growing costs. Ejection fraction (EF) is a key metric for the diagnosis and management of heart failure however estimation of EF using echocardiography remains expensive for the healthcare system and subject to intra/inter operator variability. While chest x-rays (CXR) are quick, inexpensive, and require less expertise, they do not provide sufficient information to the human eye to estimate EF. This work explores the efficacy of computer vision techniques to predict reduced EF solely from CXRs. We studied a dataset of 3488 CXRs from the MIMIC CXR-jpg (MCR) dataset. Our work establishes benchmarks using multiple state-of-the-art convolutional neural network architectures. The subsequent analysis shows increasing model sizes from 8M to 23M parameters improved classification performance without overfitting the dataset. We further show how data augmentation techniques such as CXR rotation and random cropping further improves model performance another ~5%. Finally, we conduct an error analysis using saliency maps and Grad-CAMs to better understand the failure modes of convolutional models on this task.