For Automatic Speech Recognition (ASR), the CTC-based methods have become a dominant paradigm due to its simple architecture and efficient non-autoregressive inference manner. However, these methods without external language models usually lack the capacity of modeling the conditional dependencies and the textual interaction. In this work, we present a Gated Interlayer Collaboration (GIC) mechanism which introduces the contextual information into the models and relaxes the conditional independence assumption of the CTC-based models. Specifically, we train the model with intermediate CTC losses calculated by the interlayer outputs of the model, in which the probability distributions of the intermediate layers naturally serve as soft label sequences. The GIC block consists of an embedding layer to obtain the textual embedding of the soft label at each position, and a gate unit to fuse the textual embedding and the acoustic features. Experiments on AISHELL-1 and AIDATATANG benchmarks show that the proposed method outperforms the recently published CTC-based ASR models. Specifically, our method achieves CER of 4.0%/4.4% on AISHELL-1 dev/test sets and CER of 3.8%/4.4% on AIDATATANG dev/test sets using CTC greedy search decoding without external language models.
The choice of modeling units affects the performance of the acoustic modeling and plays an important role in automatic speech recognition (ASR). In mandarin scenarios, the Chinese characters represent meaning but are not directly related to the pronunciation. Thus only considering the writing of Chinese characters as modeling units is insufficient to capture speech features. In this paper, we present a novel method involves with multi-level modeling units, which integrates multi-level information for mandarin speech recognition. Specifically, the encoder block considers syllables as modeling units, and the decoder block deals with character modeling units. During inference, the input feature sequences are converted into syllable sequences by the encoder block and then converted into Chinese characters by the decoder block. This process is conducted by a unified end-to-end model without introducing additional conversion models. By introducing InterCE auxiliary task, our method achieves competitive results with CER of 4.1%/4.6% and 4.6%/5.2% on the widely used AISHELL-1 benchmark without a language model, using the Conformer and the Transformer backbones respectively.
Dynamic attention mechanism and global modeling ability make Transformer show strong feature learning ability. In recent years, Transformer has become comparable to CNNs methods in computer vision. This review mainly investigates the current research progress of Transformer in image and video applications, which makes a comprehensive overview of Transformer in visual learning understanding. First, the attention mechanism is reviewed, which plays an essential part in Transformer. And then, the visual Transformer model and the principle of each module are introduced. Thirdly, the existing Transformer-based models are investigated, and their performance is compared in visual learning understanding applications. Three image tasks and two video tasks of computer vision are investigated. The former mainly includes image classification, object detection, and image segmentation. The latter contains object tracking and video classification. It is significant for comparing different models' performance in various tasks on several public benchmark data sets. Finally, ten general problems are summarized, and the developing prospects of the visual Transformer are given in this review.
Recent years have seen the wide application of NLP models in crucial areas such as finance, medical treatment, and news media, raising concerns of the model robustness and vulnerabilities. In this paper, we propose a novel prompt-based adversarial attack to compromise NLP models and robustness enhancement technique. We first construct malicious prompts for each instance and generate adversarial examples via mask-and-filling under the effect of a malicious purpose. Our attack technique targets the inherent vulnerabilities of NLP models, allowing us to generate samples even without interacting with the victim NLP model, as long as it is based on pre-trained language models (PLMs). Furthermore, we design a prompt-based adversarial training method to improve the robustness of PLMs. As our training method does not actually generate adversarial samples, it can be applied to large-scale training sets efficiently. The experimental results show that our attack method can achieve a high attack success rate with more diverse, fluent and natural adversarial examples. In addition, our robustness enhancement method can significantly improve the robustness of models to resist adversarial attacks. Our work indicates that prompting paradigm has great potential in probing some fundamental flaws of PLMs and fine-tuning them for downstream tasks.
Collecting dialogue state labels, slots and values, for learning dialogue state tracking (DST) models can be costly, especially with the wide application of dialogue systems in new-rising domains. In this paper, we focus on how to learn a DST model efficiently with limited labeled data. We design a prompt learning framework for few-shot DST, which consists of two main components: value-based prompt and inverse prompt mechanism. This framework aims to utilize the language understanding and generation ability of pre-trained language models (PLM). First, we design value-based prompt functions to probe the DST-related knowledge from PLM, which do not rely on the known ontology of slots. Further, an inverse prompt mechanism is utilized to self-check the "prompted" knowledge and help the PLM understand the essence of DST task further. Experiments show that our model can generate unseen slots and outperforms existing state-of-the-art few-shot methods. It indicates that DST-related knowledge can be probed from PLM and utilized to address low-resource DST efficiently with the help of prompt learning.
Deep-learning-based NLP models are found to be vulnerable to word substitution perturbations. Before they are widely adopted, the fundamental issues of robustness need to be addressed. Along this line, we propose a formal framework to evaluate word-level robustness. First, to study safe regions for a model, we introduce robustness radius which is the boundary where the model can resist any perturbation. As calculating the maximum robustness radius is computationally hard, we estimate its upper and lower bound. We repurpose attack methods as ways of seeking upper bound and design a pseudo-dynamic programming algorithm for a tighter upper bound. Then verification method is utilized for a lower bound. Further, for evaluating the robustness of regions outside a safe radius, we reexamine robustness from another view: quantification. A robustness metric with a rigorous statistical guarantee is introduced to measure the quantification of adversarial examples, which indicates the model's susceptibility to perturbations outside the safe radius. The metric helps us figure out why state-of-the-art models like BERT can be easily fooled by a few word substitutions, but generalize well in the presence of real-world noises.
This paper introduces the system submitted by the Yidun NISP team to the video keyword wakeup challenge. We propose a mandarin keyword spotting system (KWS) with several novel and effective improvements, including a big backbone (B) model, a keyword biasing (B) mechanism and the introduction of syllable modeling units (S). By considering this, we term the total system BBS-KWS as an abbreviation. The BBS-KWS system consists of an end-to-end automatic speech recognition (ASR) module and a KWS module. The ASR module converts speech features to text representations, which applies a big backbone network to the acoustic model and takes syllable modeling units into consideration as well. In addition, the keyword biasing mechanism is used to improve the recall rate of keywords in the ASR inference stage. The KWS module applies multiple criteria to determine the absence or presence of the keywords, such as multi-stage matching, fuzzy matching, and connectionist temporal classification (CTC) prefix score. To further improve our system, we conduct semi-supervised learning on the CN-Celeb dataset for better generalization. In the VKW task, the BBS-KWS system achieves significant gains over the baseline and won the first place in two tracks.
This paper introduces a notation of $\varepsilon$-weakened robustness for analyzing the reliability and stability of deep neural networks (DNNs). Unlike the conventional robustness, which focuses on the "perfect" safe region in the absence of adversarial examples, $\varepsilon$-weakened robustness focuses on the region where the proportion of adversarial examples is bounded by user-specified $\varepsilon$. Smaller $\varepsilon$ means a smaller chance of failure. Under such robustness definition, we can give conclusive results for the regions where conventional robustness ignores. We prove that the $\varepsilon$-weakened robustness decision problem is PP-complete and give a statistical decision algorithm with user-controllable error bound. Furthermore, we derive an algorithm to find the maximum $\varepsilon$-weakened robustness radius. The time complexity of our algorithms is polynomial in the dimension and size of the network. So, they are scalable to large real-world networks. Besides, We also show its potential application in analyzing quality issues.
Geohazards such as landslides have caused great losses to the safety of people's lives and property, which is often accompanied with surface cracks. If such surface cracks could be identified in time, it is of great significance for the monitoring and early warning of geohazards. Currently, the most common method for crack identification is manual detection, which is with low efficiency and accuracy. In this paper, a deep transfer learning framework is proposed to effectively and efficiently identify slope surface cracks for the sake of fast monitoring and early warning of geohazards such as landslides. The essential idea is to employ transfer learning by training (a) the large sample dataset of concrete cracks and (b) the small sample dataset of soil and rock masses cracks. In the proposed framework, (1) pretrained cracks identification models are constructed based on the large sample dataset of concrete cracks; (2) refined cracks identification models are further constructed based on the small sample dataset of soil and rock masses cracks. The proposed framework could be applied to conduct UAV surveys on high-steep slopes to realize the monitoring and early warning of landslides to ensure the safety of people's lives and property.