虽然现代自动语音识别（ASR）系统可以实现高性能，但它们可能会产生削弱读者体验并对下游任务造成伤害的错误。为了提高ASR假设的准确性和可靠性，我们提出了一种用于语音识别器的跨模型后处理系统，其中1）熔断来自不同方式的声学特征和文本特征，2）接合置信度估计器和多个误差校正器任务学习时尚和3）统一纠错和话语抑制模块。与单模或单任务模型相比，我们提出的系统被证明更有效和高效。实验结果表明，我们的后处理系统导致对工业ASR系统的单扬声器和多扬声器语音相对降低的10％相对减少，每个令牌约为1.7ms延迟确保在流语音识别中可以接受后处理引入的额外延迟。

为了提高实例级别检测/分割性能，现有的自我监督和半监督方法从未标记的数据提取非常任务 - 无关或非常任务特定的训练信号。我们认为这两种方法在任务特异性频谱的两端是任务性能的次优。利用太少的任务特定的培训信号导致底下地区任务的地面真理标签导致磨损，而相反的原因会在地面真理标签上过度装修。为此，我们提出了一种新的类别无关的半监督预测（CASP）框架，在提取来自未标记数据的训练信号中实现更有利的任务特异性平衡。与半监督学习相比，CASP通过忽略伪标签中的类信息并具有仅使用任务 - 不相关的未标记数据的单独预先预订阶段来减少训练信号的任务特异性。另一方面，CASP通过利用盒子/面具级伪标签来保留适量的任务特异性。因此，我们的预磨模模型可以更好地避免在下游任务上的FineTuned时避免在地面真理标签上抵抗/过度拟合。使用3.6M未标记的数据，我们在对象检测上实现了4.7％的显着性能增益。我们的预制模型还展示了对其他检测和分割任务/框架的优异可转移性。

Large training data and expensive model tweaking are standard features of deep learning for images. As a result, data owners often utilize cloud resources to develop large-scale complex models, which raises privacy concerns. Existing solutions are either too expensive to be practical or do not sufficiently protect the confidentiality of data and models. In this paper, we study and compare novel \emph{image disguising} mechanisms, DisguisedNets and InstaHide, aiming to achieve a better trade-off among the level of protection for outsourced DNN model training, the expenses, and the utility of data. DisguisedNets are novel combinations of image blocktization, block-level random permutation, and two block-level secure transformations: random multidimensional projection (RMT) and AES pixel-level encryption (AES). InstaHide is an image mixup and random pixel flipping technique \cite{huang20}. We have analyzed and evaluated them under a multi-level threat model. RMT provides a better security guarantee than InstaHide, under the Level-1 adversarial knowledge with well-preserved model quality. In contrast, AES provides a security guarantee under the Level-2 adversarial knowledge, but it may affect model quality more. The unique features of image disguising also help us to protect models from model-targeted attacks. We have done an extensive experimental evaluation to understand how these methods work in different settings for different datasets.

A storyboard is a roadmap for video creation which consists of shot-by-shot images to visualize key plots in a text synopsis. Creating video storyboards however remains challenging which not only requires association between high-level texts and images, but also demands for long-term reasoning to make transitions smooth across shots. In this paper, we propose a new task called Text synopsis to Video Storyboard (TeViS) which aims to retrieve an ordered sequence of images to visualize the text synopsis. We construct a MovieNet-TeViS benchmark based on the public MovieNet dataset. It contains 10K text synopses each paired with keyframes that are manually selected from corresponding movies by considering both relevance and cinematic coherence. We also present an encoder-decoder baseline for the task. The model uses a pretrained vision-and-language model to improve high-level text-image matching. To improve coherence in long-term shots, we further propose to pre-train the decoder on large-scale movie frames without text. Experimental results demonstrate that our proposed model significantly outperforms other models to create text-relevant and coherent storyboards. Nevertheless, there is still a large gap compared to human performance suggesting room for promising future work.

Solving real-world optimal control problems are challenging tasks, as the system dynamics can be highly non-linear or including nonconvex objectives and constraints, while in some cases the dynamics are unknown, making it hard to numerically solve the optimal control actions. To deal with such modeling and computation challenges, in this paper, we integrate Neural Networks with the Pontryagin's Minimum Principle (PMP), and propose a computationally efficient framework NN-PMP. The resulting controller can be implemented for systems with unknown and complex dynamics. It can not only utilize the accurate surrogate models parameterized by neural networks, but also efficiently recover the optimality conditions along with the optimal action sequences via PMP conditions. A toy example on a nonlinear Martian Base operation along with a real-world lossy energy storage arbitrage example demonstrates our proposed NN-PMP is a general and versatile computation tool for finding optimal solutions. Compared with solutions provided by the numerical optimization solver with approximated linear dynamics, NN-PMP achieves more efficient system modeling and higher performance in terms of control objectives.

The task of reconstructing 3D human motion has wideranging applications. The gold standard Motion capture (MoCap) systems are accurate but inaccessible to the general public due to their cost, hardware and space constraints. In contrast, monocular human mesh recovery (HMR) methods are much more accessible than MoCap as they take single-view videos as inputs. Replacing the multi-view Mo- Cap systems with a monocular HMR method would break the current barriers to collecting accurate 3D motion thus making exciting applications like motion analysis and motiondriven animation accessible to the general public. However, performance of existing HMR methods degrade when the video contains challenging and dynamic motion that is not in existing MoCap datasets used for training. This reduces its appeal as dynamic motion is frequently the target in 3D motion recovery in the aforementioned applications. Our study aims to bridge the gap between monocular HMR and multi-view MoCap systems by leveraging information shared across multiple video instances of the same action. We introduce the Neural Motion (NeMo) field. It is optimized to represent the underlying 3D motions across a set of videos of the same action. Empirically, we show that NeMo can recover 3D motion in sports using videos from the Penn Action dataset, where NeMo outperforms existing HMR methods in terms of 2D keypoint detection. To further validate NeMo using 3D metrics, we collected a small MoCap dataset mimicking actions in Penn Action,and show that NeMo achieves better 3D reconstruction compared to various baselines.

A major goal of multimodal research is to improve machine understanding of images and text. Tasks include image captioning, text-to-image generation, and vision-language representation learning. So far, research has focused on the relationships between images and text. For example, captioning models attempt to understand the semantics of images which are then transformed into text. An important question is: which annotation reflects best a deep understanding of image content? Similarly, given a text, what is the best image that can present the semantics of the text? In this work, we argue that the best text or caption for a given image is the text which would generate the image which is the most similar to that image. Likewise, the best image for a given text is the image that results in the caption which is best aligned with the original text. To this end, we propose a unified framework that includes both a text-to-image generative model and an image-to-text generative model. Extensive experiments validate our approach.

Model-based attacks can infer training data information from deep neural network models. These attacks heavily depend on the attacker's knowledge of the application domain, e.g., using it to determine the auxiliary data for model-inversion attacks. However, attackers may not know what the model is used for in practice. We propose a generative adversarial network (GAN) based method to explore likely or similar domains of a target model -- the model domain inference (MDI) attack. For a given target (classification) model, we assume that the attacker knows nothing but the input and output formats and can use the model to derive the prediction for any input in the desired form. Our basic idea is to use the target model to affect a GAN training process for a candidate domain's dataset that is easy to obtain. We find that the target model may distract the training procedure less if the domain is more similar to the target domain. We then measure the distraction level with the distance between GAN-generated datasets, which can be used to rank candidate domains for the target model. Our experiments show that the auxiliary dataset from an MDI top-ranked domain can effectively boost the result of model-inversion attacks.

To reproduce the success of text-to-image (T2I) generation, recent works in text-to-video (T2V) generation employ large-scale text-video dataset for fine-tuning. However, such paradigm is computationally expensive. Humans have the amazing ability to learn new visual concepts from just one single exemplar. We hereby study a new T2V generation problem$\unicode{x2014}$One-Shot Video Generation, where only a single text-video pair is presented for training an open-domain T2V generator. Intuitively, we propose to adapt the T2I diffusion model pretrained on massive image data for T2V generation. We make two key observations: 1) T2I models are able to generate images that align well with the verb terms; 2) extending T2I models to generate multiple images concurrently exhibits surprisingly good content consistency. To further learn continuous motion, we propose Tune-A-Video with a tailored Sparse-Causal Attention, which generates videos from text prompts via an efficient one-shot tuning of pretrained T2I diffusion models. Tune-A-Video is capable of producing temporally-coherent videos over various applications such as change of subject or background, attribute editing, style transfer, demonstrating the versatility and effectiveness of our method.

Detecting actions in untrimmed videos should not be limited to a small, closed set of classes. We present a simple, yet effective strategy for open-vocabulary temporal action detection utilizing pretrained image-text co-embeddings. Despite being trained on static images rather than videos, we show that image-text co-embeddings enable openvocabulary performance competitive with fully-supervised models. We show that the performance can be further improved by ensembling the image-text features with features encoding local motion, like optical flow based features, or other modalities, like audio. In addition, we propose a more reasonable open-vocabulary evaluation setting for the ActivityNet data set, where the category splits are based on similarity rather than random assignment.