Zhongqin Wu

Ye Yuan, Xiao Liu, Wondimu Dikubab et al.

Handwritten mathematical expression recognition (HMER) is a challenging task that has many potential applications. Recent methods for HMER have achieved outstanding performance with an encoder-decoder architecture. However, these methods adhere to the paradigm that the prediction is made "from one character to another", which inevitably yields prediction errors due to the complicated structures of mathematical expressions or crabbed handwritings. In this paper, we propose a simple and efficient method for HMER, which is the first to incorporate syntax information into an encoder-decoder network. Specifically, we present a set of grammar rules for converting the LaTeX markup sequence of each expression into a parsing tree; then, we model the markup sequence prediction as a tree traverse process with a deep neural network. In this way, the proposed method can effectively describe the syntax context of expressions, alleviating the structure prediction errors of HMER. Experiments on three benchmark datasets demonstrate that our method achieves better recognition performance than prior arts. To further validate the effectiveness of our method, we create a large-scale dataset consisting of 100k handwritten mathematical expression images acquired from ten thousand writers. The source code, new dataset, and pre-trained models of this work will be publicly available.

Yupeng Shi, Xiao Liu, Yuxiang Wei et al.

Semantic image synthesis is a challenging task with many practical applications. Albeit remarkable progress has been made in semantic image synthesis with spatially-adaptive normalization and existing methods normalize the feature activations under the coarse-level guidance (e.g., semantic class). However, different parts of a semantic object (e.g., wheel and window of car) are quite different in structures and textures, making blurry synthesis results usually inevitable due to the missing of fine-grained guidance. In this paper, we propose a novel normalization module, termed as REtrieval-based Spatially AdaptIve normaLization (RESAIL), for introducing pixel level fine-grained guidance to the normalization architecture. Specifically, we first present a retrieval paradigm by finding a content patch of the same semantic class from training set with the most similar shape to each test semantic mask. Then, RESAIL is presented to use the retrieved patch for guiding the feature normalization of corresponding region, and can provide pixel level fine-grained guidance, thereby greatly mitigating blurry synthesis results. Moreover, distorted ground-truth images are also utilized as alternatives of retrieval-based guidance for feature normalization, further benefiting model training and improving visual quality of generated images. Experiments on several challenging datasets show that our RESAIL performs favorably against state-of-the-arts in terms of quantitative metrics, visual quality, and subjective evaluation. The source code and pre-trained models will be publicly available.

Xingchen Song, Di Wu, Binbin Zhang et al.

The recently proposed Conformer architecture which combines convolution with attention to capture both local and global dependencies has become the \textit{de facto} backbone model for Automatic Speech Recognition~(ASR). Inherited from the Natural Language Processing (NLP) tasks, the architecture takes Layer Normalization~(LN) as a default normalization technique. However, through a series of systematic studies, we find that LN might take 10\% of the inference time despite that it only contributes to 0.1\% of the FLOPs. This motivates us to replace LN with other normalization techniques, e.g., Batch Normalization~(BN), to speed up inference with the help of operator fusion methods and the avoidance of calculating the mean and variance statistics during inference. After examining several plain attempts which directly remove all LN layers or replace them with BN in the same place, we find that the divergence issue is mainly caused by the unstable layer output. We therefore propose to append a BN layer to each linear or convolution layer where stabilized training results are observed. We also propose to simplify the activations in Conformer, such as Swish and GLU, by replacing them with ReLU. All these exchanged modules can be fused into the weights of the adjacent linear/convolution layers and hence have zero inference cost. Therefore, we name it FusionFormer. Our experiments indicate that FusionFormer is as effective as the LN-based Conformer and is about 10\% faster.

Yuxiang Wei, Yupeng Shi, Xiao Liu et al.

Unsupervised disentanglement learning is a crucial issue for understanding and exploiting deep generative models. Recently, SeFa tries to find latent disentangled directions by performing SVD on the first projection of a pre-trained GAN. However, it is only applied to the first layer and works in a post-processing way. Hessian Penalty minimizes the off-diagonal entries of the output's Hessian matrix to facilitate disentanglement, and can be applied to multi-layers.However, it constrains each entry of output independently, making it not sufficient in disentangling the latent directions (e.g., shape, size, rotation, etc.) of spatially correlated variations. In this paper, we propose a simple Orthogonal Jacobian Regularization (OroJaR) to encourage deep generative model to learn disentangled representations. It simply encourages the variation of output caused by perturbations on different latent dimensions to be orthogonal, and the Jacobian with respect to the input is calculated to represent this variation. We show that our OroJaR also encourages the output's Hessian matrix to be diagonal in an indirect manner. In contrast to the Hessian Penalty, our OroJaR constrains the output in a holistic way, making it very effective in disentangling latent dimensions corresponding to spatially correlated variations. Quantitative and qualitative experimental results show that our method is effective in disentangled and controllable image generation, and performs favorably against the state-of-the-art methods. Our code is available at https://github.com/csyxwei/OroJaR

Shiting Xu, Guowei Xu, Peilei Jia et al.

Task requirements (TRs) writing is an important question type in Key English Test and Preliminary English Test. A TR writing question may include multiple requirements and a high-quality essay must respond to each requirement thoroughly and accurately. However, the limited teacher resources prevent students from getting detailed grading instantly. The majority of existing automatic essay scoring systems focus on giving a holistic score but rarely provide reasons to support it. In this paper, we proposed an end-to-end framework based on machine reading comprehension (MRC) to address this problem to some extent. The framework not only detects whether an essay responds to a requirement question, but clearly marks where the essay answers the question. Our framework consists of three modules: question normalization module, ELECTRA based MRC module and response locating module. We extensively explore state-of-the-art MRC methods. Our approach achieves 0.93 accuracy score and 0.85 F1 score on a real-world educational dataset. To encourage reproducible results, we make our code publicly available at \url{https://github.com/aied2021TRMRC/AIED_2021_TRMRC_code}.

Hang Li, Yu Kang, Yang Hao et al.

The quality of vocal delivery is one of the key indicators for evaluating teacher enthusiasm, which has been widely accepted to be connected to the overall course qualities. However, existing evaluation for vocal delivery is mainly conducted with manual ratings, which faces two core challenges: subjectivity and time-consuming. In this paper, we present a novel machine learning approach that utilizes pairwise comparisons and a multimodal orthogonal fusing algorithm to generate large-scale objective evaluation results of the teacher vocal delivery in terms of fluency and passion. We collect two datasets from real-world education scenarios and the experiment results demonstrate the effectiveness of our algorithm. To encourage reproducible results, we make our code public available at \url{https://github.com/tal-ai/ML4VocalDelivery.git}.

Qiongqiong Liu, Tianqiao Liu, Jiafu Zhao et al.

Sentence completion (SC) questions present a sentence with one or more blanks that need to be filled in, three to five possible words or phrases as options. SC questions are widely used for students learning English as a Second Language (ESL) and building computational approaches to automatically solve such questions is beneficial to language learners. In this work, we propose a neural framework to solve SC questions in English examinations by utilizing pre-trained language models. We conduct extensive experiments on a real-world K-12 ESL SC question dataset and the results demonstrate the superiority of our model in terms of prediction accuracy. Furthermore, we run precision-recall trade-off analysis to discuss the practical issues when deploying it in real-life scenarios. To encourage reproducible results, we make our code publicly available at \url{https://github.com/AIED2021/ESL-SentenceCompletion}.

Yang Hao, Hang Li, Wenbiao Ding et al.

In this work, we study computational approaches to detect online dialogic instructions, which are widely used to help students understand learning materials, and build effective study habits. This task is rather challenging due to the widely-varying quality and pedagogical styles of dialogic instructions. To address these challenges, we utilize pre-trained language models, and propose a multi-task paradigm which enhances the ability to distinguish instances of different classes by enlarging the margin between categories via contrastive loss. Furthermore, we design a strategy to fully exploit the misclassified examples during the training stage. Extensive experiments on a real-world online educational data set demonstrate that our approach achieves superior performance compared to representative baselines. To encourage reproducible results, we make our implementation online available at \url{https://github.com/AIED2021/multitask-dialogic-instruction}.

Guowei Xu, Wenbiao Ding, Weiping Fu et al.

Many real-world applications involve the use of Optical Character Recognition (OCR) engines to transform handwritten images into transcripts on which downstream Natural Language Processing (NLP) models are applied. In this process, OCR engines may introduce errors and inputs to downstream NLP models become noisy. Despite that pre-trained models achieve state-of-the-art performance in many NLP benchmarks, we prove that they are not robust to noisy texts generated by real OCR engines. This greatly limits the application of NLP models in real-world scenarios. In order to improve model performance on noisy OCR transcripts, it is natural to train the NLP model on labelled noisy texts. However, in most cases there are only labelled clean texts. Since there is no handwritten pictures corresponding to the text, it is impossible to directly use the recognition model to obtain noisy labelled data. Human resources can be employed to copy texts and take pictures, but it is extremely expensive considering the size of data for model training. Consequently, we are interested in making NLP models intrinsically robust to OCR errors in a low resource manner. We propose a novel robust training framework which 1) employs simple but effective methods to directly simulate natural OCR noises from clean texts and 2) iteratively mines the hard examples from a large number of simulated samples for optimal performance. 3) To make our model learn noise-invariant representations, a stability loss is employed. Experiments on three real-world datasets show that the proposed framework boosts the robustness of pre-trained models by a large margin. We believe that this work can greatly promote the application of NLP models in actual scenarios, although the algorithm we use is simple and straightforward. We make our codes and three datasets publicly available\footnote{https://github.com/tal-ai/Robust-learning-MSSHEM}.

Hang Li, Wenbiao Ding, Zhongqin Wu et al.

Speech emotion recognition is a challenging task because the emotion expression is complex, multimodal and fine-grained. In this paper, we propose a novel multimodal deep learning approach to perform fine-grained emotion recognition from real-life speeches. We design a temporal alignment mean-max pooling mechanism to capture the subtle and fine-grained emotions implied in every utterance. In addition, we propose a cross modality excitement module to conduct sample-specific adjustment on cross modality embeddings and adaptively recalibrate the corresponding values by its aligned latent features from the other modality. Our proposed model is evaluated on two well-known real-world speech emotion recognition datasets. The results demonstrate that our approach is superior on the prediction tasks for multimodal speech utterances, and it outperforms a wide range of baselines in terms of prediction accuracy. Further more, we conduct detailed ablation studies to show that our temporal alignment mean-max pooling mechanism and cross modality excitement significantly contribute to the promising results. In order to encourage the research reproducibility, we make the code publicly available at \url{https://github.com/tal-ai/FG_CME.git}.

Tianqiao Liu, Qiang Fang, Wenbiao Ding et al.

There is an increasing interest in the use of mathematical word problem (MWP) generation in educational assessment. Different from standard natural question generation, MWP generation needs to maintain the underlying mathematical operations between quantities and variables, while at the same time ensuring the relevance between the output and the given topic. To address above problem, we develop an end-to-end neural model to generate diverse MWPs in real-world scenarios from commonsense knowledge graph and equations. The proposed model (1) learns both representations from edge-enhanced Levi graphs of symbolic equations and commonsense knowledge; (2) automatically fuses equation and commonsense knowledge information via a self-planning module when generating the MWPs. Experiments on an educational gold-standard set and a large-scale generated MWP set show that our approach is superior on the MWP generation task, and it outperforms the SOTA models in terms of both automatic evaluation metrics, i.e., BLEU-4, ROUGE-L, Self-BLEU, and human evaluation metrics, i.e., equation relevance, topic relevance, and language coherence. To encourage reproducible results, we make our code and MWP dataset public available at \url{https://github.com/tal-ai/MaKE_EMNLP2021}.

Xijun Wang, Dongyang Liu, Meina Kan et al.

In Knowledge Distillation, the teacher is generally much larger than the student, making the solution of the teacher likely to be difficult for the student to learn. To ease the mimicking difficulty, we introduce a triplet knowledge distillation mechanism named TriKD. Besides teacher and student, TriKD employs a third role called anchor model. Before distillation begins, the pre-trained anchor model delimits a subspace within the full solution space of the target problem. Solutions within the subspace are expected to be easy targets that the student could mimic well. Distillation then begins in an online manner, and the teacher is only allowed to express solutions within the aforementioned subspace. Surprisingly, benefiting from accurate but easy-to-mimic hints, the student can finally perform well. After the student is well trained, it can be used as the new anchor for new students, forming a curriculum learning strategy. Our experiments on image classification and face recognition with various models clearly demonstrate the effectiveness of our method. Furthermore, the proposed TriKD is also effective in dealing with the overfitting issue. Moreover, our theoretical analysis supports the rationality of our triplet distillation.

Xueyuan Chen, Changhe Song, Yixuan Zhou et al.

The accuracy of prosodic structure prediction is crucial to the naturalness of synthesized speech in Mandarin text-to-speech system, but now is limited by widely-used sequence-to-sequence framework and error accumulation from previous word segmentation results. In this paper, we propose a span-based Mandarin prosodic structure prediction model to obtain an optimal prosodic structure tree, which can be converted to corresponding prosodic label sequence. Instead of the prerequisite for word segmentation, rich linguistic features are provided by Chinese character-level BERT and sent to encoder with self-attention architecture. On top of this, span representation and label scoring are used to describe all possible prosodic structure trees, of which each tree has its corresponding score. To find the optimal tree with the highest score for a given sentence, a bottom-up CKY-style algorithm is further used. The proposed method can predict prosodic labels of different levels at the same time and accomplish the process directly from Chinese characters in an end-to-end manner. Experiment results on two real-world datasets demonstrate the excellent performance of our span-based method over all sequence-to-sequence baseline approaches.

Han Xu, Xiaorui Liu, Wentao Wang et al.

Recent studies suggest that ``memorization'' is one important factor for overparameterized deep neural networks (DNNs) to achieve optimal performance. Specifically, the perfectly fitted DNNs can memorize the labels of many atypical samples, generalize their memorization to correctly classify test atypical samples and enjoy better test performance. While, DNNs which are optimized via adversarial training algorithms can also achieve perfect training performance by memorizing the labels of atypical samples, as well as the adversarially perturbed atypical samples. However, adversarially trained models always suffer from poor generalization, with both relatively low clean accuracy and robustness on the test set. In this work, we study the effect of memorization in adversarial trained DNNs and disclose two important findings: (a) Memorizing atypical samples is only effective to improve DNN's accuracy on clean atypical samples, but hardly improve their adversarial robustness and (b) Memorizing certain atypical samples will even hurt the DNN's performance on typical samples. Based on these two findings, we propose Benign Adversarial Training (BAT) which can facilitate adversarial training to avoid fitting ``harmful'' atypical samples and fit as more ``benign'' atypical samples as possible. In our experiments, we validate the effectiveness of BAT, and show it can achieve better clean accuracy vs. robustness trade-off than baseline methods, in benchmark datasets such as CIFAR100 and Tiny~ImageNet.

Lu Ma, Song Yang, Yaguang Gong et al.

Acoustic Echo Cancellation (AEC) plays a key role in speech interaction by suppressing the echo received at microphone introduced by acoustic reverberations from loudspeakers. Since the performance of linear adaptive filter (AF) would degrade severely due to nonlinear distortions, background noises, and microphone clipping in real scenarios, deep learning has been employed for AEC for its good nonlinear modelling ability. In this paper, we constructed an end-to-end multi-scale attention neural network for AEC. Temporal convolution is first used to transform waveform into spectrogram. The spectrograms of the far-end reference and the near-end mixture are concatenated, and fed to a temporal convolution network (TCN) with stacked dilated convolution layers. Attention mechanism is performed among these representations from different layers to adaptively extract relevant features by referring to the previous hidden state in the encoder long short-term memory (LSTM) unit. The representations are weighted averaged and fed to the encoder LSTM for the near-end speech estimation. Experiments show the superiority of our method in terms of the echo return loss enhancement (ERLE) for single-talk periods and the perceptual evaluation of speech quality (PESQ) score for double-talk periods in background noise and nonlinear distortion scenarios.

Lu Ma, Song Yang, Yaguang Gong et al.

This paper proposes an noise type classification aided attention-based neural network approach for monaural speech enhancement. The network is constructed based on a previous work by introducing a noise classification subnetwork into the structure and taking the classification embedding into the attention mechanism for guiding the network to make better feature extraction. Specifically, to make the network an end-to-end way, an audio encoder and decoder constructed by temporal convolution is used to make transformation between waveform and spectrogram. Additionally, our model is composed of two long short term memory (LSTM) based encoders, two attention mechanism, a noise classifier and a speech mask generator. Experiments show that, compared with OM-LSA and the previous work, the proposed noise classification aided attention-based approach can achieve better performance in terms of speech quality (PESQ). More promisingly, our approach has better generalization ability to unseen noise conditions.

Lu Ma, Xintian Wang, Song Yang et al.

Teaching with the cooperation of expert teacher and assistant teacher, which is the so-called "double-teachers classroom", i.e., the course is giving by the expert online and presented through projection screen at the classroom, and the teacher at the classroom performs as an assistant for guiding the students in learning, is becoming more prevalent in today's teaching method for K-12 education. For monitoring the teaching quality, a microphone clipped on the assistant's neckline is always used for voice recording, then fed to the downstream tasks of automatic speech recognition (ASR) and neural language processing (NLP). However, besides its voice, there would be some other interfering voices, including the expert's one and the student's one. Here, we propose to extract the assistant' voices from the perspective of sound event detection, i.e., the voices are classified into four categories, namely the expert, the teacher, the mixture of them, and the background. To make frame-level identification, which is important for grabbing sensitive words for the downstream tasks, a multi-scale temporal convolution neural network is constructed with stacked dilated convolutions for considering both local and global properties. These features are concatenated and fed to a classification network constructed by three linear layers. The framework is evaluated on simulated data and real-world recordings, giving considerable performance in terms of precision and recall, compared with some classical classification methods.

Lu Ma, Song Yang, Yaguang Gong et al.

Acoustic Echo Cancellation (AEC) whose aim is to suppress the echo originated from acoustic coupling between loudspeakers and microphones, plays a key role in voice interaction. Linear adaptive filter (AF) is always used for handling this problem. However, since there would be some severe effects in real scenarios, such nonlinear distortions, background noises, and microphone clipping, it would lead to considerable residual echo, giving poor performance in practice. In this paper, we propose an end-to-end network structure for echo cancellation, which is directly done on time-domain audio waveform. It is transformed to deep representation by temporal convolution, and modelled by Long Short-Term Memory (LSTM) for considering temporal property. Since time delay and severe reverberation may exist at the near-end with respect to the far-end, a local attention is employed for alignment. The network is trained using multitask learning by employing an auxiliary classification network for double-talk detection. Experiments show the superiority of our proposed method in terms of the echo return loss enhancement (ERLE) for single-talk periods and the perceptual evaluation of speech quality (PESQ) score for double-talk periods in background noise and nonlinear distortion scenarios.

Yue Cao, Xiaohe Wu, Shuran Qi et al.

The success of deep denoisers on real-world color photographs usually relies on the modeling of sensor noise and in-camera signal processing (ISP) pipeline. Performance drop will inevitably happen when the sensor and ISP pipeline of test images are different from those for training the deep denoisers (i.e., noise discrepancy). In this paper, we present an unpaired learning scheme to adapt a color image denoiser for handling test images with noise discrepancy. We consider a practical training setting, i.e., a pre-trained denoiser, a set of test noisy images, and an unpaired set of clean images. To begin with, the pre-trained denoiser is used to generate the pseudo clean images for the test images. Pseudo-ISP is then suggested to jointly learn the pseudo ISP pipeline and signal-dependent rawRGB noise model using the pairs of test and pseudo clean images. We further apply the learned pseudo ISP and rawRGB noise model to clean color images to synthesize realistic noisy images for denoiser adaption. Pseudo-ISP is effective in synthesizing realistic noisy sRGB images, and improved denoising performance can be achieved by alternating between Pseudo-ISP training and denoiser adaption. Experiments show that our Pseudo-ISP not only can boost simple Gaussian blurring-based denoiser to achieve competitive performance against CBDNet, but also is effective in improving state-of-the-art deep denoisers, e.g., CBDNet and RIDNet.

Zhihong Shao, Zitao Liu, Jiyong Zhang et al.

Adversarial examples are vital to expose the vulnerability of machine learning models. Despite the success of the most popular substitution-based methods which substitutes some characters or words in the original examples, only substitution is insufficient to uncover all robustness issues of models. In this paper, we present AdvExpander, a method that crafts new adversarial examples by expanding text, which is complementary to previous substitution-based methods. We first utilize linguistic rules to determine which constituents to expand and what types of modifiers to expand with. We then expand each constituent by inserting an adversarial modifier searched from a CVAE-based generative model which is pre-trained on a large scale corpus. To search adversarial modifiers, we directly search adversarial latent codes in the latent space without tuning the pre-trained parameters. To ensure that our adversarial examples are label-preserving for text matching, we also constrain the modifications with a heuristic rule. Experiments on three classification tasks verify the effectiveness of AdvExpander and the validity of our adversarial examples. AdvExpander crafts a new type of adversarial examples by text expansion, thereby promising to reveal new robustness issues.

Haochen Liu, Zitao Liu, Zhongqin Wu et al.

The automatic evaluation for school assignments is an important application of AI in the education field. In this work, we focus on the task of personalized multimodal feedback generation, which aims to generate personalized feedback for various teachers to evaluate students' assignments involving multimodal inputs such as images, audios, and texts. This task involves the representation and fusion of multimodal information and natural language generation, which presents the challenges from three aspects: 1) how to encode and integrate multimodal inputs; 2) how to generate feedback specific to each modality; and 3) how to realize personalized feedback generation. In this paper, we propose a novel Personalized Multimodal Feedback Generation Network (PMFGN) armed with a modality gate mechanism and a personalized bias mechanism to address these challenges. The extensive experiments on real-world K-12 education data show that our model significantly outperforms several baselines by generating more accurate and diverse feedback. In addition, detailed ablation experiments are conducted to deepen our understanding of the proposed framework.

Pengcheng Wang, Zihao Wang, Zhilong Ji et al.

This paper introduces our approach to the EmotioNet Challenge 2020. We pose the AU recognition problem as a multi-task learning problem, where the non-rigid facial muscle motion (mainly the first 17 AUs) and the rigid head motion (the last 6 AUs) are modeled separately. The co-occurrence of the expression features and the head pose features are explored. We observe that different AUs converge at various speed. By choosing the optimal checkpoint for each AU, the recognition results are improved. We are able to obtain a final score of 0.746 in validation set and 0.7306 in the test set of the challenge.