Neural Architecture Search (NAS) is an automatic technique that can search for well-performed architectures for a specific task. Although NAS surpasses human-designed architecture in many fields, the high computational cost of architecture evaluation it requires hinders its development. A feasible solution is to directly evaluate some metrics in the initial stage of the architecture without any training. NAS without training (WOT) score is such a metric, which estimates the final trained accuracy of the architecture through the ability to distinguish different inputs in the activation layer. However, WOT score is not an atomic metric, meaning that it does not represent a fundamental indicator of the architecture. The contributions of this paper are in three folds. First, we decouple WOT into two atomic metrics which represent the distinguishing ability of the network and the number of activation units, and explore better combination rules named (Distinguishing Activation Score) DAS. We prove the correctness of decoupling theoretically and confirmed the effectiveness of the rules experimentally. Second, in order to improve the prediction accuracy of DAS to meet practical search requirements, we propose a fast training strategy. When DAS is used in combination with the fast training strategy, it yields more improvements. Third, we propose a dataset called Darts-training-bench (DTB), which fills the gap that no training states of architecture in existing datasets. Our proposed method has 1.04$\times$ - 1.56$\times$ improvements on NAS-Bench-101, Network Design Spaces, and the proposed DTB.

Homography estimation is erroneous in the case of large-baseline due to the low image overlay and limited receptive field. To address it, we propose a progressive estimation strategy by converting large-baseline homography into multiple intermediate ones, cumulatively multiplying these intermediate items can reconstruct the initial homography. Meanwhile, a semi-supervised homography identity loss, which consists of two components: a supervised objective and an unsupervised objective, is introduced. The first supervised loss is acting to optimize intermediate homographies, while the second unsupervised one helps to estimate a large-baseline homography without photometric losses. To validate our method, we propose a large-scale dataset that covers regular and challenging scenes. Experiments show that our method achieves state-of-the-art performance in large-baseline scenes while keeping competitive performance in small-baseline scenes. Code and dataset are available at https://github.com/megvii-research/LBHomo.

Dialogue state tracking (DST) aims to convert the dialogue history into dialogue states which consist of slot-value pairs. As condensed structural information memorizing all history information, the dialogue state in the last turn is typically adopted as the input for predicting the current state by DST models. However, these models tend to keep the predicted slot values unchanged, which is defined as state momentum in this paper. Specifically, the models struggle to update slot values that need to be changed and correct wrongly predicted slot values in the last turn. To this end, we propose MoNET to tackle state momentum via noise-enhanced training. First, the previous state of each turn in the training data is noised via replacing some of its slot values. Then, the noised previous state is used as the input to learn to predict the current state, improving the model's ability to update and correct slot values. Furthermore, a contrastive context matching framework is designed to narrow the representation distance between a state and its corresponding noised variant, which reduces the impact of noised state and makes the model better understand the dialogue history. Experimental results on MultiWOZ datasets show that MoNET outperforms previous DST methods. Ablations and analysis verify the effectiveness of MoNET in alleviating state momentum and improving anti-noise ability.

Video super-resolution is one of the most popular tasks on mobile devices, being widely used for an automatic improvement of low-bitrate and low-resolution video streams. While numerous solutions have been proposed for this problem, they are usually quite computationally demanding, demonstrating low FPS rates and power efficiency on mobile devices. In this Mobile AI challenge, we address this problem and propose the participants to design an end-to-end real-time video super-resolution solution for mobile NPUs optimized for low energy consumption. The participants were provided with the REDS training dataset containing video sequences for a 4X video upscaling task. The runtime and power efficiency of all models was evaluated on the powerful MediaTek Dimensity 9000 platform with a dedicated AI processing unit capable of accelerating floating-point and quantized neural networks. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 500 FPS rate and 0.2 [Watt / 30 FPS] power consumption. A detailed description of all models developed in the challenge is provided in this paper.

本文衡量了跨语言寄存器变化的稳定性。寄存器是各种与语言上下文相关的语言。寄存器及其上下文之间的关系是功能的：构成寄存器的语言特征是由交流状况的需求和约束所激发的。该观点假设寄存器应该是通用的，因此我们期望定义寄存器的语言外部环境与寄存器所包含的语言特征集之间存在稳定的关系。在本文中，使用在可比的交流情况下生成的Corpora在60种语言中比较寄存器特定语言中的变化来测试寄存器变化的普遍性和鲁棒性：推文和Wikipedia文章。我们的发现证实了寄存器变化实际上是普遍的预测。

本文使用寄存器预测任务进行了39种语言的基于频率语料库相似性的实验。目的是量化（i）不同语料库与同一语言和（ii）单个语音的同质性之间的距离。这两个目标对于衡量基于语料库的语言分析如何从一个数据集推广到另一个数据集都至关重要。问题在于，以前的工作集中在印欧语上，提出了一个问题，即这些措施是否能够在各种语言上提供强大的概括。本文使用寄存器预测任务来评估跨39种语言的竞争措施：他们能够区分代表不同生产环境的语料库？每个实验都将单个语言的三个语料库与所有语言共享的三个数字寄存器进行比较：社交媒体，网页和Wikipedia。结果表明，语料库相似性的衡量标准保留了不同语言家族，写作系统和形态类型的有效性。此外，当对不域外的语料库，应用于低资源语言以及应用于不同的寄存器集时，这些措施仍然坚固。鉴于我们需要在可用于分析的迅速增加的情况下进行概括，因此这些发现很重要。

本文模拟了17种语言的低资源设置，以评估不同条件下的相似性，稳定性和可靠性。目的是在训练之前使用语料库相似性度量，以预测训练后嵌入的特性。本文的主要贡献是表明可以使用上游语料库相似性度量来预测下游嵌入相似性。然后，通过建模从非常有限的训练数据创建的嵌入式的可靠性，将此发现应用于低资源设置。结果表明，可以使用语料库相似性度量估算低资源嵌入的可靠性，这些度量在少量数据上保持强大。这些发现对评估真正低资源语言的评估具有重大影响，在这种语言中，由于数据限制，这种系统的下游验证方法是不可能的。

Most existing text-video retrieval methods focus on cross-modal matching between the visual content of offline videos and textual query sentences. However, in real scenarios, online videos are frequently accompanied by relevant text information such as titles, tags, and even subtitles, which can be utilized to match textual queries. This inspires us to generate associated captions from offline videos to help with existing text-video retrieval methods. To do so, we propose to use the zero-shot video captioner with knowledge of pre-trained web-scale models (e.g., CLIP and GPT-2) to generate captions for offline videos without any training. Given the captions, one question naturally arises: what can auxiliary captions do for text-video retrieval? In this paper, we present a novel framework Cap4Video, which makes use of captions from three aspects: i) Input data: The video and captions can form new video-caption pairs as data augmentation for training. ii) Feature interaction: We perform feature interaction between video and caption to yield enhanced video representations. iii) Output score: The Query-Caption matching branch can be complementary to the original Query-Video matching branch for text-video retrieval. We conduct thorough ablation studies to demonstrate the effectiveness of our method. Without any post-processing, our Cap4Video achieves state-of-the-art performance on MSR-VTT (51.4%), VATEX (66.6%), MSVD (51.8%), and DiDeMo (52.0%).

Vision-language models (VLMs) that are pre-trained on large-scale image-text pairs have demonstrated impressive transferability on a wide range of visual tasks. Transferring knowledge from such powerful pre-trained VLMs is emerging as a promising direction for building effective video recognition models. However, the current exploration is still limited. In our opinion, the greatest charm of pre-trained vision-language models is to build a bridge between visual and textual domains. In this paper, we present a novel framework called BIKE which utilizes the cross-modal bridge to explore bidirectional knowledge: i) We propose a Video Attribute Association mechanism which leverages the Video-to-Text knowledge to generate textual auxiliary attributes to complement video recognition. ii) We also present a Temporal Concept Spotting mechanism which uses the Text-to-Video expertise to capture temporal saliency in a parameter-free manner to yield enhanced video representation. The extensive studies on popular video datasets (ie, Kinetics-400 & 600, UCF-101, HMDB-51 and ActivityNet) show that our method achieves state-of-the-art performance in most recognition scenarios, eg, general, zero-shot, and few-shot video recognition. To the best of our knowledge, our best model achieves a state-of-the-art accuracy of 88.4% on challenging Kinetics-400 with the released CLIP pre-trained model.

Frost damage is one of the main factors leading to wheat yield reduction. Therefore, the detection of wheat frost accurately and efficiently is beneficial for growers to take corresponding measures in time to reduce economic loss. To detect the wheat frost, in this paper we create a hyperspectral wheat frost data set by collecting the data characterized by temperature, wheat yield, and hyperspectral information provided by the handheld hyperspectral spectrometer. However, due to the imbalance of data, that is, the number of healthy samples is much higher than the number of frost damage samples, a deep learning algorithm tends to predict biasedly towards the healthy samples resulting in model overfitting of the healthy samples. Therefore, we propose a method based on deep cost-sensitive learning, which uses a one-dimensional convolutional neural network as the basic framework and incorporates cost-sensitive learning with fixed factors and adjustment factors into the loss function to train the network. Meanwhile, the accuracy and score are used as evaluation metrics. Experimental results show that the detection accuracy and the score reached 0.943 and 0.623 respectively, this demonstration shows that this method not only ensures the overall accuracy but also effectively improves the detection rate of frost samples.