We present a graph convolutional network with 2D pose estimation for the first time on child action recognition task achieving on par results with an RGB modality based model on a novel benchmark dataset containing unconstrained environment based videos.

This paper presents an implementation on child activity recognition (CAR) with a graph convolution network (GCN) based deep learning model since prior implementations in this domain have been dominated by CNN, LSTM and other methods despite the superior performance of GCN. To the best of our knowledge, we are the first to use a GCN model in child activity recognition domain. In overcoming the challenges of having small size publicly available child action datasets, several learning methods such as feature extraction, fine-tuning and curriculum learning were implemented to improve the model performance. Inspired by the contradicting claims made on the use of transfer learning in CAR, we conducted a detailed implementation and analysis on transfer learning together with a study on negative transfer learning effect on CAR as it hasn't been addressed previously. As the principal contribution, we were able to develop a ST-GCN based CAR model which, despite the small size of the dataset, obtained around 50% accuracy on vanilla implementations. With feature extraction and fine-tuning methods, accuracy was improved by 20%-30% with the highest accuracy being 82.24%. Furthermore, the results provided on activity datasets empirically demonstrate that with careful selection of pre-train model datasets through methods such as curriculum learning could enhance the accuracy levels. Finally, we provide preliminary evidence on possible frame rate effect on the accuracy of CAR models, a direction future research can explore.

Despite recent progress in Natural Language Understanding (NLU), the creation of multilingual NLU systems remains a challenge. It is common to have NLU systems limited to a subset of languages due to lack of available data. They also often vary widely in performance. We launch a three-phase approach to address the limitations in NLU and help propel NLU technology to new heights. We release a 52 language dataset called the Multilingual Amazon SLU resource package (SLURP) for Slot-filling, Intent classification, and Virtual assistant Evaluation, or MASSIVE, in an effort to address parallel data availability for voice assistants. We organize the Massively Multilingual NLU 2022 Challenge to provide a competitive environment and push the state-of-the art in the transferability of models into other languages. Finally, we host the first Massively Multilingual NLU workshop which brings these components together. The MMNLU workshop seeks to advance the science behind multilingual NLU by providing a platform for the presentation of new research in the field and connecting teams working on this research direction. This paper summarizes the dataset, workshop and the competition and the findings of each phase.

在这项工作中，我们证明了多种语的大规模序列到序列（SEQ2SEQ）模型，该模型是通过Denoising和因果语言建模（CLM）任务的混合物进行训练的，比仅解码器模型更有效地进行了效率的学习者在各种任务上。特别是，我们培训了一个名为Alexa教师模型（Alexatm 20b）的200亿个参数多语言SEQ2SEQ模型，并表明它在1-Shot摘要任务上实现了最先进的（SOTA）性能，超过了更大的540B PALM DOPODER模型。 Alexatm 20b还可以在1-Shot Machine翻译中实现SOTA，尤其是对于低资源语言，几乎所有语言对（阿拉伯语，英语，法语，德语，德语，印地语，意大利语，日语，以及flores-101数据集上的泰卢固语）。我们还显示了零拍设置，AlexATM 20B在SuperGlue和SqueadV2数据集上的表现优于GPT3（175B），并在XNLI，XCOPA，PAWS-X和XWINOGRAD等多语言任务上提供SOTA性能。总体而言，我们的结果为SEQ2SEQ模型提供了一个令人信服的案例，作为大型语言模型（LLM）培训的仅解码器模型的强大替代方法。

我们介绍了一个大规模实验，该实验对编码器进行了预处理，其参数计数范围从700m到9.3b不等，随后蒸馏到较小的型号中，范围为17m-170亿参数，其应用到自然语言理解（NLU）组件（NLU）组件（虚拟助手系统。尽管我们使用70％的口语数据训练，但在对书面形式的跨语性自然语言推论（XNLI）语料库进行评估时，我们的教师模型与XLM-R和MT5相当。我们使用系统中的内域数据对教师模型进行了第二阶段的训练，以提高了3.86％的相对分类，而相对7.01％的插槽填充。我们发现，即使是从我们的2阶段教师模型中提取的170亿参数模型，与仅接受公共数据的2.3B参数老师相比，与2.3B参数老师相比，意图分类更好2.88％，并且7.69％的插槽填充错误率更好（第1阶段），强调了。内域数据对训练的重要性。当使用标记的NLU数据进行离线评估时，我们的17m参数阶段2蒸馏模型的表现分别优于XLM-R碱基（85m Params）和Distillbert（42m Params），分别优于4.23％至6.14％。最后，我们介绍了一个完整的虚拟助手实验平台的结果，在该平台中，我们发现使用经过预训练和蒸馏管道训练的模型超过了从8500万参数教师蒸馏的模型，在自动测量全系统用户不满的自动测量中，从8500万参数教师蒸馏出3.74％-4.91％。

最近的大规模自然语言处理（NLP）系统对大规模和多样化的语料库使用预先培训的大型语言模型（LLM）。实际上，预训练的模型通过对特定于任务的数据集进行微调来适应各种任务。 LLMS虽然有效，但已被证明可以记住培训数据的实例，从而有可能揭示在预训练期间处理的私人信息。潜在的泄漏可能会进一步传播到LLM经过微调的下游任务。另一方面，保存隐私的算法通常涉及从头开始的重新划痕，这对于LLM来说非常昂贵。在这项工作中，我们提出了一个简单，易于解释的，并且在解码阶段将其轻巧的扰动机制应用于已经训练的模型。我们的扰动机制是模型不可抑制的，可以与任何LLM结合使用。我们提供的理论分析表明，所提出的机制是私人的，实验结果显示了隐私 - 私人权衡权衡。

我们介绍了用于插槽，意图分类和虚拟助手评估的大规模数据集 - 数字亚马逊SLU资源包（SLURP）。大规模包含1M现实，平行，标记为虚拟助手的话语，涵盖51种语言，18个域，60个意图和55个插槽。通过任务专业翻译人员将仅英文slurp数据集定位为29属的50种类型多样性的语言来创建大规模。我们还介绍了XLM-R和MT5上的建模结果，包括精确的匹配精度，意图分类精度和插槽填充F1分数。我们已经公开发布了数据集，建模代码和模型。