Biao Li

Tianyi Xu, Zhanheng Yang, Kaixun Huang et al.

By incorporating additional contextual information, deep biasing methods have emerged as a promising solution for speech recognition of personalized words. However, for real-world voice assistants, always biasing on such personalized words with high prediction scores can significantly degrade the performance of recognizing common words. To address this issue, we propose an adaptive contextual biasing method based on Context-Aware Transformer Transducer (CATT) that utilizes the biased encoder and predictor embeddings to perform streaming prediction of contextual phrase occurrences. Such prediction is then used to dynamically switch the bias list on and off, enabling the model to adapt to both personalized and common scenarios. Experiments on Librispeech and internal voice assistant datasets show that our approach can achieve up to 6.7% and 20.7% relative reduction in WER and CER compared to the baseline respectively, mitigating up to 96.7% and 84.9% of the relative WER and CER increase for common cases. Furthermore, our approach has a minimal performance impact in personalized scenarios while maintaining a streaming inference pipeline with negligible RTF increase.

Ruigang Fu, Qingyong Hu, Xiaohu Dong et al.

As a fundamental operation in modern machine vision models, feature upsampling has been widely used and investigated in the literatures. An ideal upsampling operation should be lightweight, with low computational complexity. That is, it can not only improve the overall performance but also not affect the model complexity. Content-aware Reassembly of Features (CARAFE) is a well-designed learnable operation to achieve feature upsampling. Albeit encouraging performance achieved, this method requires generating large-scale kernels, which brings a mass of extra redundant parameters, and inherently has limited scalability. To this end, we propose a lightweight upsampling operation, termed Dynamic Lightweight Upsampling (DLU) in this paper. In particular, it first constructs a small-scale source kernel space, and then samples the large-scale kernels from the kernel space by introducing learnable guidance offsets, hence avoiding introducing a large collection of trainable parameters in upsampling. Experiments on several mainstream vision tasks show that our DLU achieves comparable and even better performance to the original CARAFE, but with much lower complexity, e.g., DLU requires 91% fewer parameters and at least 63% fewer FLOPs (Floating Point Operations) than CARAFE in the case of 16x upsampling, but outperforms the CARAFE by 0.3% mAP in object detection. Code is available at https://github.com/Fu0511/Dynamic-Lightweight-Upsampling.

Xiaoming Ren, Chao Li, Shenjian Wang et al.

Considering the bimodal nature of human speech perception, lips, and teeth movement has a pivotal role in automatic speech recognition. Benefiting from the correlated and noise-invariant visual information, audio-visual recognition systems enhance robustness in multiple scenarios. In previous work, audio-visual HuBERT appears to be the finest practice incorporating modality knowledge. This paper outlines a mixed methodology, named conformer enhanced AV-HuBERT, boosting the AV-HuBERT system's performance a step further. Compared with baseline AV-HuBERT, our method in the one-phase evaluation of clean and noisy conditions achieves 7% and 16% relative WER reduction on the English AVSR benchmark dataset LRS3. Furthermore, we establish a novel 1000h Mandarin AVSR dataset CSTS. On top of the baseline AV-HuBERT, we exceed the WeNet ASR system by 14% and 18% relatively on MISP and CMLR by pre-training with this dataset. The conformer-enhanced AV-HuBERT we proposed brings 7% on MISP and 6% CER reduction on CMLR, compared with the baseline AV-HuBERT system.

Jia Wang, Ping Wang, Biao Li et al.

Remote sensing image registration is valuable for image-based navigation system despite posing many challenges. As the search space of registration is usually non-convex, the optimization algorithm, which aims to search the best transformation parameters, is a challenging step. Conventional optimization algorithms can hardly reconcile the contradiction of simultaneous rapid convergence and the global optimization. In this paper, a novel learning-based optimization algorithm named Image Registration Optimizer Network (IRON) is proposed, which can predict the global optimum after single iteration. The IRON is trained by a 3D tensor (9x9x9), which consists of similar metric values. The elements of the 3D tensor correspond to the 9x9x9 neighbors of the initial parameters in the search space. Then, the tensor's label is a vector that points to the global optimal parameters from the initial parameters. Because of the special architecture, the IRON could predict the global optimum directly for any initialization. The experimental results demonstrate that the proposed algorithm performs better than other classical optimization algorithms as it has higher accuracy, lower root of mean square error (RMSE), and more efficiency. Our IRON codes are available for further study.https://www.github.com/jaxwangkd04/IRON

Ruigang Fu, Qingyong Hu, Xiaohu Dong et al.

To have a better understanding and usage of Convolution Neural Networks (CNNs), the visualization and interpretation of CNNs has attracted increasing attention in recent years. In particular, several Class Activation Mapping (CAM) methods have been proposed to discover the connection between CNN's decision and image regions. In spite of the reasonable visualization, lack of clear and sufficient theoretical support is the main limitation of these methods. In this paper, we introduce two axioms -- Conservation and Sensitivity -- to the visualization paradigm of the CAM methods. Meanwhile, a dedicated Axiom-based Grad-CAM (XGrad-CAM) is proposed to satisfy these axioms as much as possible. Experiments demonstrate that XGrad-CAM is an enhanced version of Grad-CAM in terms of conservation and sensitivity. It is able to achieve better visualization performance than Grad-CAM, while also be class-discriminative and easy-to-implement compared with Grad-CAM++ and Ablation-CAM. The code is available at https://github.com/Fu0511/XGrad-CAM.

Meng Ai, Biao Li, Heyang Gong et al.

Offering incentives (e.g., coupons at Amazon, discounts at Uber and video bonuses at Tiktok) to user is a common strategy used by online platforms to increase user engagement and platform revenue. Despite its proven effectiveness, these marketing incentives incur an inevitable cost and might result in a low ROI (Return on Investment) if not used properly. On the other hand, different users respond differently to these incentives, for instance, some users never buy certain products without coupons, while others do anyway. Thus, how to select the right amount of incentives (i.e. treatment) to each user under budget constraints is an important research problem with great practical implications. In this paper, we call such problem as a budget-constrained treatment selection (BTS) problem. The challenge is how to efficiently solve BTS problem on a Large-Scale dataset and achieve improved results over the existing techniques. We propose a novel tree-based treatment selection technique under budget constraints, called Large-Scale Budget-Constrained Causal Forest (LBCF) algorithm, which is also an efficient treatment selection algorithm suitable for modern distributed computing systems. A novel offline evaluation method is also proposed to overcome an intrinsic challenge in assessing solutions' performance for BTS problem in randomized control trials (RCT) data. We deploy our approach in a real-world scenario on a large-scale video platform, where the platform gives away bonuses in order to increase users' campaign engagement duration. The simulation analysis, offline and online experiments all show that our method outperforms various tree-based state-of-the-art baselines. The proposed approach is currently serving over hundreds of millions of users on the platform and achieves one of the most tremendous improvements over these months.

Biao Li, Jiabin Liu, Bo Wang et al.

Deep learning (DL) architectures for superresolution (SR) normally contain tremendous parameters, which has been regarded as the crucial advantage for obtaining satisfying performance. However, with the widespread use of mobile phones for taking and retouching photos, this character greatly hampers the deployment of DL-SR models on the mobile devices. To address this problem, in this paper, we propose a super lightweight SR network: s-LWSR. There are mainly three contributions in our work. Firstly, in order to efficiently abstract features from the low resolution image, we build an information pool to mix multi-level information from the first half part of the pipeline. Accordingly, the information pool feeds the second half part with the combination of hierarchical features from the previous layers. Secondly, we employ a compression module to further decrease the size of parameters. Intensive analysis confirms its capacity of trade-off between model complexity and accuracy. Thirdly, by revealing the specific role of activation in deep models, we remove several activation layers in our SR model to retain more information for performance improvement. Extensive experiments show that our s-LWSR, with limited parameters and operations, can achieve similar performance to other cumbersome DL-SR methods.

Yaxiong Wang, Hao Yang, Xueming Qian et al.

Image-text matching tasks have recently attracted a lot of attention in the computer vision field. The key point of this cross-domain problem is how to accurately measure the similarity between the visual and the textual contents, which demands a fine understanding of both modalities. In this paper, we propose a novel position focused attention network (PFAN) to investigate the relation between the visual and the textual views. In this work, we integrate the object position clue to enhance the visual-text joint-embedding learning. We first split the images into blocks, by which we infer the relative position of region in the image. Then, an attention mechanism is proposed to model the relations between the image region and blocks and generate the valuable position feature, which will be further utilized to enhance the region expression and model a more reliable relationship between the visual image and the textual sentence. Experiments on the popular datasets Flickr30K and MS-COCO show the effectiveness of the proposed method. Besides the public datasets, we also conduct experiments on our collected practical large-scale news dataset (Tencent-News) to validate the practical application value of proposed method. As far as we know, this is the first attempt to test the performance on the practical application. Our method achieves the state-of-art performance on all of these three datasets.