Despite the popularity of Vision Transformers (ViTs) and eXplainable AI (XAI), only a few explanation methods have been proposed for ViTs thus far. They use attention weights of the classification token on patch embeddings and often produce unsatisfactory saliency maps. In this paper, we propose a novel method for explaining ViTs called ViT-CX. It is based on patch embeddings, rather than attentions paid to them, and their causal impacts on the model output. ViT-CX can be used to explain different ViT models. Empirical results show that, in comparison with previous methods, ViT-CX produces more meaningful saliency maps and does a better job at revealing all the important evidence for prediction. It is also significantly more faithful to the model as measured by deletion AUC and insertion AUC.

从概念上讲，深层聚类模型由一个特征提取器组成，该功能提取器将数据映射到潜在空间，以及将数据指向潜在空间中簇的聚类头。尽管这两个组件曾经以端到端的方式共同培训，但最近的作品证明，在两个阶段分别训练它们是有益的。在第一阶段，特征提取器通过自我监督的学习进行训练，从而可以保留数据点之间的群集结构。为了更好地保留群集结构，我们建议通过通过自我监督的学习在更大的数据集上鉴定的另一个模型来代替第一阶段。该方法很简单，可能会遭受域转移的困扰。尽管如此，我们从经验上表明，它可以实现出色的聚类性能。当视觉变压器（VIT）结构用于特征提取时，我们的方法分别达到了CIFAR-10，CIFAR-100和STL-10的聚类精度94.0％，55.6％和97.9％。相应的先前最新结果为84.3％，47.7％和80.8％。我们的代码将在本文发布后在线提供。

在只有有限的数据可用的低资源场景中，自然语言处理（NLP）的建立模型（NLP）具有挑战性。基于优化的元学习算法通过适应良好的模型初始化来处理新任务，从而在低资源场景中实现了有希望的结果。尽管如此，这些方法遭受了记忆过度拟合问题的困扰，在这种情况下，模型倾向于记住元训练任务，而在适应新任务时忽略了支持集。为了解决此问题，我们提出了一种内存模仿元学习（MEMIML）方法，该方法增强了模型对任务适应的支持集的依赖。具体来说，我们引入了一个特定于任务的内存模块来存储支持集信息并构建一个模仿模块，以强制查询集，以模仿存储在存储器中的某些代表性支持集样本的行为。提供了一种理论分析来证明我们方法的有效性，经验结果还表明，我们的方法在文本分类和生成任务上都优于竞争基准。

The proliferation of automatic faithfulness metrics for summarization has produced a need for benchmarks to evaluate them. While existing benchmarks measure the correlation with human judgements of faithfulness on model-generated summaries, they are insufficient for diagnosing whether metrics are: 1) consistent, i.e., decrease as errors are introduced into a summary, 2) effective on human-written texts, and 3) sensitive to different error types (as summaries can contain multiple errors). To address these needs, we present a benchmark of unfaithful minimal pairs (BUMP), a dataset of 889 human-written, minimally different summary pairs, where a single error (from an ontology of 7 types) is introduced to a summary from the CNN/DailyMail dataset to produce an unfaithful summary. We find BUMP complements existing benchmarks in a number of ways: 1) the summaries in BUMP are harder to discriminate and less probable under SOTA summarization models, 2) BUMP enables measuring the consistency of metrics, and reveals that the most discriminative metrics tend not to be the most consistent, 3) BUMP enables the measurement of metrics' performance on individual error types and highlights areas of weakness for future work.

The number of international benchmarking competitions is steadily increasing in various fields of machine learning (ML) research and practice. So far, however, little is known about the common practice as well as bottlenecks faced by the community in tackling the research questions posed. To shed light on the status quo of algorithm development in the specific field of biomedical imaging analysis, we designed an international survey that was issued to all participants of challenges conducted in conjunction with the IEEE ISBI 2021 and MICCAI 2021 conferences (80 competitions in total). The survey covered participants' expertise and working environments, their chosen strategies, as well as algorithm characteristics. A median of 72% challenge participants took part in the survey. According to our results, knowledge exchange was the primary incentive (70%) for participation, while the reception of prize money played only a minor role (16%). While a median of 80 working hours was spent on method development, a large portion of participants stated that they did not have enough time for method development (32%). 25% perceived the infrastructure to be a bottleneck. Overall, 94% of all solutions were deep learning-based. Of these, 84% were based on standard architectures. 43% of the respondents reported that the data samples (e.g., images) were too large to be processed at once. This was most commonly addressed by patch-based training (69%), downsampling (37%), and solving 3D analysis tasks as a series of 2D tasks. K-fold cross-validation on the training set was performed by only 37% of the participants and only 50% of the participants performed ensembling based on multiple identical models (61%) or heterogeneous models (39%). 48% of the respondents applied postprocessing steps.

The security of artificial intelligence (AI) is an important research area towards safe, reliable, and trustworthy AI systems. To accelerate the research on AI security, the Artificial Intelligence Security Competition (AISC) was organized by the Zhongguancun Laboratory, China Industrial Control Systems Cyber Emergency Response Team, Institute for Artificial Intelligence, Tsinghua University, and RealAI as part of the Zhongguancun International Frontier Technology Innovation Competition (https://www.zgc-aisc.com/en). The competition consists of three tracks, including Deepfake Security Competition, Autonomous Driving Security Competition, and Face Recognition Security Competition. This report will introduce the competition rules of these three tracks and the solutions of top-ranking teams in each track.

End-to-end multilingual ASR has become more appealing because of several reasons such as simplifying the training and deployment process and positive performance transfer from high-resource to low-resource languages. However, scaling up the number of languages, total hours, and number of unique tokens is not a trivial task. This paper explores large-scale multilingual ASR models on 70 languages. We inspect two architectures: (1) Shared embedding and output and (2) Multiple embedding and output model. In the shared model experiments, we show the importance of tokenization strategy across different languages. Later, we use our optimal tokenization strategy to train multiple embedding and output model to further improve our result. Our multilingual ASR achieves 13.9%-15.6% average WER relative improvement compared to monolingual models. We show that our multilingual ASR generalizes well on an unseen dataset and domain, achieving 9.5% and 7.5% WER on Multilingual Librispeech (MLS) with zero-shot and finetuning, respectively.

Despite the huge advancement in knowledge discovery and data mining techniques, the X-ray diffraction (XRD) analysis process has mostly remained untouched and still involves manual investigation, comparison, and verification. Due to the large volume of XRD samples from high-throughput XRD experiments, it has become impossible for domain scientists to process them manually. Recently, they have started leveraging standard clustering techniques, to reduce the XRD pattern representations requiring manual efforts for labeling and verification. Nevertheless, these standard clustering techniques do not handle problem-specific aspects such as peak shifting, adjacent peaks, background noise, and mixed phases; hence, resulting in incorrect composition-phase diagrams that complicate further steps. Here, we leverage data mining techniques along with domain expertise to handle these issues. In this paper, we introduce an incremental phase mapping approach based on binary peak representations using a new threshold based fuzzy dissimilarity measure. The proposed approach first applies an incremental phase computation algorithm on discrete binary peak representation of XRD samples, followed by hierarchical clustering or manual merging of similar pure phases to obtain the final composition-phase diagram. We evaluate our method on the composition space of two ternary alloy systems- Co-Ni-Ta and Co-Ti-Ta. Our results are verified by domain scientists and closely resembles the manually computed ground-truth composition-phase diagrams. The proposed approach takes us closer towards achieving the goal of complete end-to-end automated XRD analysis.

Solar activity is usually caused by the evolution of solar magnetic fields. Magnetic field parameters derived from photospheric vector magnetograms of solar active regions have been used to analyze and forecast eruptive events such as solar flares and coronal mass ejections. Unfortunately, the most recent solar cycle 24 was relatively weak with few large flares, though it is the only solar cycle in which consistent time-sequence vector magnetograms have been available through the Helioseismic and Magnetic Imager (HMI) on board the Solar Dynamics Observatory (SDO) since its launch in 2010. In this paper, we look into another major instrument, namely the Michelson Doppler Imager (MDI) on board the Solar and Heliospheric Observatory (SOHO) from 1996 to 2010. The data archive of SOHO/MDI covers more active solar cycle 23 with many large flares. However, SOHO/MDI data only has line-of-sight (LOS) magnetograms. We propose a new deep learning method, named MagNet, to learn from combined LOS magnetograms, Bx and By taken by SDO/HMI along with H-alpha observations collected by the Big Bear Solar Observatory (BBSO), and to generate vector components Bx' and By', which would form vector magnetograms with observed LOS data. In this way, we can expand the availability of vector magnetograms to the period from 1996 to present. Experimental results demonstrate the good performance of the proposed method. To our knowledge, this is the first time that deep learning has been used to generate photospheric vector magnetograms of solar active regions for SOHO/MDI using SDO/HMI and H-alpha data.

太阳耀斑，尤其是M级和X级耀斑，通常与冠状质量弹出（CMES）有关。它们是太空天气影响的最重要来源，可能会严重影响近地环境。因此，必须预测耀斑（尤其是X级），以减轻其破坏性和危险后果。在这里，我们介绍了几种统计和机器学习方法，以预测AR的耀斑指数（FI），这些方法通过考虑到一定时间间隔内的不同类耀斑的数量来量化AR的耀斑生产力。具体而言，我们的样本包括2010年5月至2017年12月在太阳能磁盘上出现的563个AR。25个磁性参数，由空中震动和磁性成像器（HMI）的太空天气HMI活性区域（Sharp）提供的太阳能动力学观测值（HMI）。 （SDO），表征了代理中存储在ARS中的冠状磁能，并用作预测因子。我们研究了这些尖锐的参数与ARS的FI与机器学习算法（样条回归）和重采样方法（合成少数群体过度采样技术，用于使用高斯噪声回归的合成少数群体过度采样技术，smogn简短）。基于既定关系，我们能够在接下来的1天内预测给定AR的FIS值。与其他4种流行的机器学习算法相比，我们的方法提高了FI预测的准确性，尤其是对于大型FI。此外，我们根据Borda Count方法从由9种不同的机器学习方法渲染的等级计算出尖锐参数的重要性。