As one of the most popular micro-mobility options, e-scooters are spreading in hundreds of big cities and college towns in the US and worldwide. In the meantime, e-scooters are also posing new challenges to traffic safety. In general, e-scooters are suggested to be ridden in bike lanes/sidewalks or share the road with cars at the maximum speed of about 15-20 mph, which is more flexible and much faster than the pedestrains and bicyclists. These features make e-scooters challenging for human drivers, pedestrians, vehicle active safety modules, and self-driving modules to see and interact. To study this new mobility option and address e-scooter riders' and other road users' safety concerns, this paper proposes a wearable data collection system for investigating the micro-level e-Scooter motion behavior in a Naturalistic road environment. An e-Scooter-based data acquisition system has been developed by integrating LiDAR, cameras, and GPS using the robot operating system (ROS). Software frameworks are developed to support hardware interfaces, sensor operation, sensor synchronization, and data saving. The integrated system can collect data continuously for hours, meeting all the requirements including calibration accuracy and capability of collecting the vehicle and e-Scooter encountering data.

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.

Recently, the success of pre-training in text domain has been fully extended to vision, audio, and cross-modal scenarios. The proposed pre-training models of different modalities are showing a rising trend of homogeneity in their model structures, which brings the opportunity to implement different pre-training models within a uniform framework. In this paper, we present TencentPretrain, a toolkit supporting pre-training models of different modalities. The core feature of TencentPretrain is the modular design. The toolkit uniformly divides pre-training models into 5 components: embedding, encoder, target embedding, decoder, and target. As almost all of common modules are provided in each component, users can choose the desired modules from different components to build a complete pre-training model. The modular design enables users to efficiently reproduce existing pre-training models or build brand-new one. We test the toolkit on text, vision, and audio benchmarks and show that it can match the performance of the original implementations.

Large language models (LLMs) have been shown to be able to perform new tasks based on a few demonstrations or natural language instructions. While these capabilities have led to widespread adoption, most LLMs are developed by resource-rich organizations and are frequently kept from the public. As a step towards democratizing this powerful technology, we present BLOOM, a 176B-parameter open-access language model designed and built thanks to a collaboration of hundreds of researchers. BLOOM is a decoder-only Transformer language model that was trained on the ROOTS corpus, a dataset comprising hundreds of sources in 46 natural and 13 programming languages (59 in total). We find that BLOOM achieves competitive performance on a wide variety of benchmarks, with stronger results after undergoing multitask prompted finetuning. To facilitate future research and applications using LLMs, we publicly release our models and code under the Responsible AI License.

We explore the capability of plain Vision Transformers (ViTs) for semantic segmentation and propose the SegVit. Previous ViT-based segmentation networks usually learn a pixel-level representation from the output of the ViT. Differently, we make use of the fundamental component -- attention mechanism, to generate masks for semantic segmentation. Specifically, we propose the Attention-to-Mask (ATM) module, in which the similarity maps between a set of learnable class tokens and the spatial feature maps are transferred to the segmentation masks. Experiments show that our proposed SegVit using the ATM module outperforms its counterparts using the plain ViT backbone on the ADE20K dataset and achieves new state-of-the-art performance on COCO-Stuff-10K and PASCAL-Context datasets. Furthermore, to reduce the computational cost of the ViT backbone, we propose query-based down-sampling (QD) and query-based up-sampling (QU) to build a Shrunk structure. With the proposed Shrunk structure, the model can save up to $40\%$ computations while maintaining competitive performance.

眼科图像和衍生物，例如视网膜神经纤维层（RNFL）厚度图对于检测和监测眼科疾病至关重要（例如，青光眼）。对于计算机辅助诊断眼疾病，关键技术是自动从眼科图像中提取有意义的特征，这些特征可以揭示与功能视觉丧失相关的生物标志物（例如RNFL变薄模式）。然而，将结构性视网膜损伤与人类视力丧失联系起来的眼科图像的表示，主要是由于患者之间的解剖学变化很大。在存在图像伪像的情况下，这项任务变得更加具有挑战性，由于图像采集和自动细分，这很常见。在本文中，我们提出了一个耐伪造的无监督的学习框架，该框架称为眼科图像的学习表示。 Eyelearn具有一个伪影校正模块，可以学习可以最好地预测无伪影眼镜图像的表示形式。此外，Eyelearn采用聚类引导的对比度学习策略，以明确捕获内部和间形的亲和力。在训练过程中，图像在簇中动态组织，以形成对比样品，其中鼓励在相同或不同的簇中分别学习相似或不同的表示形式。为了评估包冰者，我们使用青光眼患者的现实世界眼科摄影图数据集使用学习的表示形式进行视野预测和青光眼检测。广泛的实验和与最先进方法的比较验证了眼球从眼科图像中学习最佳特征表示的有效性。

我们提出了一个新的框架，用于对凸函数的差异私有优化，这些功能是任意规范$ \ normx {\ cdot} $中的Lipschitz。我们的算法基于一种正规的指数机制，该机制从密度$ \ propto \ exp（-k（f+\ mu r））$中进行样品，其中$ f $是经验损失，$ r $是一种常规化器，它与强烈的convex convex converize尊重$ \ normx {\ cdot} $，将\ cite {gll22}的最新作品推广到非Euclidean设置。我们表明，这种机制可以满足高斯差异隐私，并通过使用凸几何形状的本地化工具来解决DP-MER（经验风险最小化）和DP-SCO（随机凸优化）。我们的框架是第一个在一般规范空间中适用于私有凸优化的框架，并直接恢复了镜下下降的非私有SCO率，作为隐私参数$ \ eps \ to \ infty $。作为应用程序，对于LipsChitz优化了$ \ ell_p $ norms for（1，2）$中的所有$ p \ norms，我们获得了第一个最佳隐私性权衡权衡；对于$ p = 1 $，我们提高了最近的作品\ cite {asifkt21，bassilygn21}获得的权衡，至少通过对数因素。我们的$ \ ell_p $ norm和schatten- $ p $规范优化框架与多项式时间采样器相辅相成，我们的查询复杂性明确绑定。

深度学习的快速发展在分割方面取得了长足的进步，这是计算机视觉的基本任务之一。但是，当前的细分算法主要取决于像素级注释的可用性，这些注释通常昂贵，乏味且费力。为了减轻这一负担，过去几年见证了越来越多的关注，以建立标签高效，深度学习的细分算法。本文对标签有效的细分方法进行了全面的审查。为此，我们首先根据不同类型的弱标签提供的监督（包括没有监督，粗略监督，不完整的监督和嘈杂的监督和嘈杂的监督），首先开发出一种分类法来组织这些方法，并通过细分类型（包括语义细分）补充，实例分割和全景分割）。接下来，我们从统一的角度总结了现有的标签有效的细分方法，该方法讨论了一个重要的问题：如何弥合弱监督和密集预测之间的差距 - 当前的方法主要基于启发式先导，例如交叉像素相似性，跨标签约束，跨视图一致性，跨图像关系等。最后，我们分享了对标签有效深层细分的未来研究方向的看法。

我们提出了一个简单而有效的完全卷积的一阶段3D对象检测器，用于自主驾驶场景的LIDAR点云，称为FCOS-LIDAR。与使用鸟眼视图（BEV）的主要方法不同，我们提出的检测器从激光雷达点的范围视图（RV，又称范围图像）中检测对象。由于范围视图的紧凑性和与LIDAR传感器在自动驾驶汽车上的采样过程的兼容性，因此可以通过仅利用Vanilla 2D卷积来实现基于范围视图的对象检测器，而脱离了基于BEV的方法，这些方法通常涉及复杂的方法体素化操作和稀疏卷积。我们首次表明，仅具有标准2D卷积的基于RV的3D检测器就可以实现与基于BEV的最新检测器相当的性能，同时更快，更简单。更重要的是，几乎所有以前的基于范围视图的检测器都只关注单帧点云，因为将多帧点云融合到单个范围视图中是具有挑战性的。在这项工作中，我们通过新颖的范围视图投影机制解决了这个具有挑战性的问题，并首次展示了基于范围视图的检测器融合多帧点云的好处。关于Nuscenes的广泛实验表明了我们提出的方法的优越性，我们认为我们的工作可以有力证明基于RV的3D检测器可以与当前基于BEV的主流探测器相比。

我们提出了一种直接的，基于回归的方法，以从单个图像中估计2D人姿势。我们将问题提出为序列预测任务，我们使用变压器网络解决了问题。该网络直接学习了从图像到关键点坐标的回归映射，而无需诉诸中间表示（例如热图）。这种方法避免了与基于热图的方法相关的许多复杂性。为了克服以前基于回归的方法的特征错位问题，我们提出了一种注意机制，该机制适应与目标关键最相关的功能，从而大大提高了准确性。重要的是，我们的框架是端到端的可区分，并且自然学会利用关键点之间的依赖关系。两个主要的姿势估计数据集在MS-Coco和MPII上进行的实验表明，我们的方法在基于回归的姿势估计中的最新方法显着改善。更值得注意的是，与最佳的基于热图的姿势估计方法相比，我们的第一种基于回归的方法是有利的。