Wei Zhou

Wei Zhou, Xiangyu Peng, Mark Riedl · gatech

One major challenge in reinforcement learning (RL) is the large amount of steps for the RL agent needs to converge in the training process and learn the optimal policy, especially in text-based game environments where the action space is extensive. However, non-player characters (NPCs) sometimes hold some key information about the game, which can potentially help to train RL agents faster. Thus, this paper explores how to interact and converse with NPC agents to get the key information using large language models (LLMs), as well as incorporate this information to speed up RL agent's training using knowledge graphs (KGs) and Story Shaping.

Xiangyu Peng, Christopher Cui, Wei Zhou et al. · gatech

Reward design for reinforcement learning agents can be difficult in situations where one not only wants the agent to achieve some effect in the world but where one also cares about how that effect is achieved. For example, we might wish for an agent to adhere to a tacit understanding of commonsense, align itself to a preference for how to behave for purposes of safety, or taking on a particular role in an interactive game. Storytelling is a mode for communicating tacit procedural knowledge. We introduce a technique, Story Shaping, in which a reinforcement learning agent infers tacit knowledge from an exemplar story of how to accomplish a task and intrinsically rewards itself for performing actions that make its current environment adhere to that of the inferred story world. Specifically, Story Shaping infers a knowledge graph representation of the world state from observations, and also infers a knowledge graph from the exemplar story. An intrinsic reward is generated based on the similarity between the agent's inferred world state graph and the inferred story world graph. We conducted experiments in text-based games requiring commonsense reasoning and shaping the behaviors of agents as virtual game characters.

Dou Hu, Yinan Bao, Lingwei Wei et al.

Extracting generalized and robust representations is a major challenge in emotion recognition in conversations (ERC). To address this, we propose a supervised adversarial contrastive learning (SACL) framework for learning class-spread structured representations in a supervised manner. SACL applies contrast-aware adversarial training to generate worst-case samples and uses joint class-spread contrastive learning to extract structured representations. It can effectively utilize label-level feature consistency and retain fine-grained intra-class features. To avoid the negative impact of adversarial perturbations on context-dependent data, we design a contextual adversarial training (CAT) strategy to learn more diverse features from context and enhance the model's context robustness. Under the framework with CAT, we develop a sequence-based SACL-LSTM to learn label-consistent and context-robust features for ERC. Experiments on three datasets show that SACL-LSTM achieves state-of-the-art performance on ERC. Extended experiments prove the effectiveness of SACL and CAT.

Wei Zhou, Wilfried Michel, Ralf Schlüter et al.

As one of the most popular sequence-to-sequence modeling approaches for speech recognition, the RNN-Transducer has achieved evolving performance with more and more sophisticated neural network models of growing size and increasing training epochs. While strong computation resources seem to be the prerequisite of training superior models, we try to overcome it by carefully designing a more efficient training pipeline. In this work, we propose an efficient 3-stage progressive training pipeline to build highly-performing neural transducer models from scratch with very limited computation resources in a reasonable short time period. The effectiveness of each stage is experimentally verified on both Librispeech and Switchboard corpora. The proposed pipeline is able to train transducer models approaching state-of-the-art performance with a single GPU in just 2-3 weeks. Our best conformer transducer achieves 4.1% WER on Librispeech test-other with only 35 epochs of training.

Yinan Bao, Qianwen Ma, Lingwei Wei et al.

The emotion recognition in conversation (ERC) task aims to predict the emotion label of an utterance in a conversation. Since the dependencies between speakers are complex and dynamic, which consist of intra- and inter-speaker dependencies, the modeling of speaker-specific information is a vital role in ERC. Although existing researchers have proposed various methods of speaker interaction modeling, they cannot explore dynamic intra- and inter-speaker dependencies jointly, leading to the insufficient comprehension of context and further hindering emotion prediction. To this end, we design a novel speaker modeling scheme that explores intra- and inter-speaker dependencies jointly in a dynamic manner. Besides, we propose a Speaker-Guided Encoder-Decoder (SGED) framework for ERC, which fully exploits speaker information for the decoding of emotion. We use different existing methods as the conversational context encoder of our framework, showing the high scalability and flexibility of the proposed framework. Experimental results demonstrate the superiority and effectiveness of SGED.

Wei Zhou, Qi Yang, Qiuping Jiang et al.

Objective quality assessment of 3D point clouds is essential for the development of immersive multimedia systems in real-world applications. Despite the success of perceptual quality evaluation for 2D images and videos, blind/no-reference metrics are still scarce for 3D point clouds with large-scale irregularly distributed 3D points. Therefore, in this paper, we propose an objective point cloud quality index with Structure Guided Resampling (SGR) to automatically evaluate the perceptually visual quality of 3D dense point clouds. The proposed SGR is a general-purpose blind quality assessment method without the assistance of any reference information. Specifically, considering that the human visual system (HVS) is highly sensitive to structure information, we first exploit the unique normal vectors of point clouds to execute regional pre-processing which consists of keypoint resampling and local region construction. Then, we extract three groups of quality-related features, including: 1) geometry density features; 2) color naturalness features; 3) angular consistency features. Both the cognitive peculiarities of the human brain and naturalness regularity are involved in the designed quality-aware features that can capture the most vital aspects of distorted 3D point clouds. Extensive experiments on several publicly available subjective point cloud quality databases validate that our proposed SGR can compete with state-of-the-art full-reference, reduced-reference, and no-reference quality assessment algorithms.

Yinan Bao, Qianwen Ma, Lingwei Wei et al.

The Emotion-Cause Pair Extraction (ECPE) task aims to extract emotions and causes as pairs from documents. We observe that the relative distance distribution of emotions and causes is extremely imbalanced in the typical ECPE dataset. Existing methods have set a fixed size window to capture relations between neighboring clauses. However, they neglect the effective semantic connections between distant clauses, leading to poor generalization ability towards position-insensitive data. To alleviate the problem, we propose a novel Multi-Granularity Semantic Aware Graph model (MGSAG) to incorporate fine-grained and coarse-grained semantic features jointly, without regard to distance limitation. In particular, we first explore semantic dependencies between clauses and keywords extracted from the document that convey fine-grained semantic features, obtaining keywords enhanced clause representations. Besides, a clause graph is also established to model coarse-grained semantic relations between clauses. Experimental results indicate that MGSAG surpasses the existing state-of-the-art ECPE models. Especially, MGSAG outperforms other models significantly in the condition of position-insensitive data.

Albert Zeyer, Robin Schmitt, Wei Zhou et al.

We introduce a novel segmental-attention model for automatic speech recognition. We restrict the decoder attention to segments to avoid quadratic runtime of global attention, better generalize to long sequences, and eventually enable streaming. We directly compare global-attention and different segmental-attention modeling variants. We develop and compare two separate time-synchronous decoders, one specifically taking the segmental nature into account, yielding further improvements. Using time-synchronous decoding for segmental models is novel and a step towards streaming applications. Our experiments show the importance of a length model to predict the segment boundaries. The final best segmental-attention model using segmental decoding performs better than global-attention, in contrast to other monotonic attention approaches in the literature. Further, we observe that the segmental model generalizes much better to long sequences of up to several minutes.

Xiu-Shen Wei, He-Yang Xu, Faen Zhang et al.

Semi-supervised few-shot learning consists in training a classifier to adapt to new tasks with limited labeled data and a fixed quantity of unlabeled data. Many sophisticated methods have been developed to address the challenges this problem comprises. In this paper, we propose a simple but quite effective approach to predict accurate negative pseudo-labels of unlabeled data from an indirect learning perspective, and then augment the extremely label-constrained support set in few-shot classification tasks. Our approach can be implemented in just few lines of code by only using off-the-shelf operations, yet it is able to outperform state-of-the-art methods on four benchmark datasets.

Wei Zhou, Haotian Wu, Jingjing Xu et al.

ASR can be improved by multi-task learning (MTL) with domain enhancing or domain adversarial training, which are two opposite objectives with the aim to increase/decrease domain variance towards domain-aware/agnostic ASR, respectively. In this work, we study how to best apply these two opposite objectives with speaker labels to improve conformer-based ASR. We also propose a novel adaptive gradient reversal layer for stable and effective adversarial training without tuning effort. Detailed analysis and experimental verification are conducted to show the optimal positions in the ASR neural network (NN) to apply speaker enhancing and adversarial training. We also explore their combination for further improvement, achieving the same performance as i-vectors plus adversarial training. Our best speaker-based MTL achieves 7\% relative improvement on the Switchboard Hub5'00 set. We also investigate the effect of such speaker-based MTL w.r.t. cleaner dataset and weaker ASR NN.

Yiting Lu, Jun Fu, Xin Li et al.

Coronary CT Angiography (CCTA) is susceptible to various distortions (e.g., artifacts and noise), which severely compromise the exact diagnosis of cardiovascular diseases. The appropriate CCTA Vessel-level Image Quality Assessment (CCTA VIQA) algorithm can be used to reduce the risk of error diagnosis. The primary challenges of CCTA VIQA are that the local part of coronary that determines final quality is hard to locate. To tackle the challenge, we formulate CCTA VIQA as a multiple-instance learning (MIL) problem, and exploit Transformer-based MIL backbone (termed as T-MIL) to aggregate the multiple instances along the coronary centerline into the final quality. However, not all instances are informative for final quality. There are some quality-irrelevant/negative instances intervening the exact quality assessment(e.g., instances covering only background or the coronary in instances is not identifiable). Therefore, we propose a Progressive Reinforcement learning based Instance Discarding module (termed as PRID) to progressively remove quality-irrelevant/negative instances for CCTA VIQA. Based on the above two modules, we propose a Reinforced Transformer Network (RTN) for automatic CCTA VIQA based on end-to-end optimization. Extensive experimental results demonstrate that our proposed method achieves the state-of-the-art performance on the real-world CCTA dataset, exceeding previous MIL methods by a large margin.

Zijian Yang, Wei Zhou, Ralf Schlüter et al.

Recently, RNN-Transducers have achieved remarkable results on various automatic speech recognition tasks. However, lattice-free sequence discriminative training methods, which obtain superior performance in hybrid models, are rarely investigated in RNN-Transducers. In this work, we propose three lattice-free training objectives, namely lattice-free maximum mutual information, lattice-free segment-level minimum Bayes risk, and lattice-free minimum Bayes risk, which are used for the final posterior output of the phoneme-based neural transducer with a limited context dependency. Compared to criteria using N-best lists, lattice-free methods eliminate the decoding step for hypotheses generation during training, which leads to more efficient training. Experimental results show that lattice-free methods gain up to 6.5% relative improvement in word error rate compared to a sequence-level cross-entropy trained model. Compared to the N-best-list based minimum Bayes risk objectives, lattice-free methods gain 40% - 70% relative training time speedup with a small degradation in performance.

Wei Zhou, Zhou Wang

There has been a growing interest in developing image super-resolution (SR) algorithms that convert low-resolution (LR) to higher resolution images, but automatically evaluating the visual quality of super-resolved images remains a challenging problem. Here we look at the problem of SR image quality assessment (SR IQA) in a two-dimensional (2D) space of deterministic fidelity (DF) versus statistical fidelity (SF). This allows us to better understand the advantages and disadvantages of existing SR algorithms, which produce images at different clusters in the 2D space of (DF, SF). Specifically, we observe an interesting trend from more traditional SR algorithms that are typically inclined to optimize for DF while losing SF, to more recent generative adversarial network (GAN) based approaches that by contrast exhibit strong advantages in achieving high SF but sometimes appear weak at maintaining DF. Furthermore, we propose an uncertainty weighting scheme based on content-dependent sharpness and texture assessment that merges the two fidelity measures into an overall quality prediction named the Super Resolution Image Fidelity (SRIF) index, which demonstrates superior performance against state-of-the-art IQA models when tested on subject-rated datasets.

Wei Zhou, Xuanlin Min, Rui Hu et al.

Real-time object detection on Unmanned Aerial Vehicles (UAVs) is a challenging issue due to the limited computing resources of edge GPU devices as Internet of Things (IoT) nodes. To solve this problem, in this paper, we propose a novel lightweight deep learning architectures named FasterX based on YOLOX model for real-time object detection on edge GPU. First, we design an effective and lightweight PixSF head to replace the original head of YOLOX to better detect small objects, which can be further embedded in the depthwise separable convolution (DS Conv) to achieve a lighter head. Then, a slimmer structure in the Neck layer termed as SlimFPN is developed to reduce parameters of the network, which is a trade-off between accuracy and speed. Furthermore, we embed attention module in the Head layer to improve the feature extraction effect of the prediction head. Meanwhile, we also improve the label assignment strategy and loss function to alleviate category imbalance and box optimization problems of the UAV dataset. Finally, auxiliary heads are presented for online distillation to improve the ability of position embedding and feature extraction in PixSF head. The performance of our lightweight models are validated experimentally on the NVIDIA Jetson NX and Jetson Nano GPU embedded platforms.Extensive experiments show that FasterX models achieve better trade-off between accuracy and latency on VisDrone2021 dataset compared to state-of-the-art models.

Wei Zhou, Ruizeng Zhang, Leida Li et al.

Image dehazing aims to restore spatial details from hazy images. There have emerged a number of image dehazing algorithms, designed to increase the visibility of those hazy images. However, much less work has been focused on evaluating the visual quality of dehazed images. In this paper, we propose a Reduced-Reference dehazed image quality evaluation approach based on Partial Discrepancy (RRPD) and then extend it to a No-Reference quality assessment metric with Blind Perception (NRBP). Specifically, inspired by the hierarchical characteristics of the human perceiving dehazed images, we introduce three groups of features: luminance discrimination, color appearance, and overall naturalness. In the proposed RRPD, the combined distance between a set of sender and receiver features is adopted to quantify the perceptually dehazed image quality. By integrating global and local channels from dehazed images, the RRPD is converted to NRBP which does not rely on any information from the references. Extensive experiment results on several dehazed image quality databases demonstrate that our proposed methods outperform state-of-the-art full-reference, reduced-reference, and no-reference quality assessment models. Furthermore, we show that the proposed dehazed image quality evaluation methods can be effectively applied to tune parameters for potential image dehazing algorithms.

Zhenpeng Su, Xing Wu, Wei Zhou et al.

Dialogue response selection aims to select an appropriate response from several candidates based on a given user and system utterance history. Most existing works primarily focus on post-training and fine-tuning tailored for cross-encoders. However, there are no post-training methods tailored for dense encoders in dialogue response selection. We argue that when the current language model, based on dense dialogue systems (such as BERT), is employed as a dense encoder, it separately encodes dialogue context and response, leading to a struggle to achieve the alignment of both representations. Thus, we propose Dial-MAE (Dialogue Contextual Masking Auto-Encoder), a straightforward yet effective post-training technique tailored for dense encoders in dialogue response selection. Dial-MAE uses an asymmetric encoder-decoder architecture to compress the dialogue semantics into dense vectors, which achieves better alignment between the features of the dialogue context and response. Our experiments have demonstrated that Dial-MAE is highly effective, achieving state-of-the-art performance on two commonly evaluated benchmarks.

Qiuping Jiang, Jiawu Xu, Yudong Mao et al.

Blind image quality assessment (BIQA), which aims to accurately predict the image quality without any pristine reference information, has been extensively concerned in the past decades. Especially, with the help of deep neural networks, great progress has been achieved. However, it remains less investigated on BIQA for night-time images (NTIs) which usually suffers from complicated authentic distortions such as reduced visibility, low contrast, additive noises, and color distortions. These diverse authentic degradations particularly challenges the design of effective deep neural network for blind NTI quality evaluation (NTIQE). In this paper, we propose a novel deep decomposition and bilinear pooling network (DDB-Net) to better address this issue. The DDB-Net contains three modules, i.e., an image decomposition module, a feature encoding module, and a bilinear pooling module. The image decomposition module is inspired by the Retinex theory and involves decoupling the input NTI into an illumination layer component responsible for illumination information and a reflection layer component responsible for content information. Then, the feature encoding module involves learning feature representations of degradations that are rooted in the two decoupled components separately. Finally, by modeling illumination-related and content-related degradations as two-factor variations, the two feature sets are bilinearly pooled together to form a unified representation for quality prediction. The superiority of the proposed DDB-Net has been well validated by extensive experiments on several benchmark datasets. The source code will be made available soon.

Zijian Yang, Wei Zhou, Ralf Schlüter et al.

Internal language model (ILM) subtraction has been widely applied to improve the performance of the RNN-Transducer with external language model (LM) fusion for speech recognition. In this work, we show that sequence discriminative training has a strong correlation with ILM subtraction from both theoretical and empirical points of view. Theoretically, we derive that the global optimum of maximum mutual information (MMI) training shares a similar formula as ILM subtraction. Empirically, we show that ILM subtraction and sequence discriminative training achieve similar effects across a wide range of experiments on Librispeech, including both MMI and minimum Bayes risk (MBR) criteria, as well as neural transducers and LMs of both full and limited context. The benefit of ILM subtraction also becomes much smaller after sequence discriminative training. We also provide an in-depth study to show that sequence discriminative training has a minimal effect on the commonly used zero-encoder ILM estimation, but a joint effect on both encoder and prediction + joint network for posterior probability reshaping including both ILM and blank suppression.

Dou Hu, Lingwei Wei, Yaxin Liu et al.

This paper describes our system designed for SemEval-2023 Task 12: Sentiment analysis for African languages. The challenge faced by this task is the scarcity of labeled data and linguistic resources in low-resource settings. To alleviate these, we propose a generalized multilingual system SACL-XLMR for sentiment analysis on low-resource languages. Specifically, we design a lexicon-based multilingual BERT to facilitate language adaptation and sentiment-aware representation learning. Besides, we apply a supervised adversarial contrastive learning technique to learn sentiment-spread structured representations and enhance model generalization. Our system achieved competitive results, largely outperforming baselines on both multilingual and zero-shot sentiment classification subtasks. Notably, the system obtained the 1st rank on the zero-shot classification subtask in the official ranking. Extensive experiments demonstrate the effectiveness of our system.

Haolan Zhan, Xuming Lin, Shaobo Cui et al.

Existing neural methods have shown great potentials towards generating informative text from structured tabular data as well as maintaining high content fidelity. However, few of them shed light on generating personalized expressions, which often requires well-aligned persona-table-text datasets that are difficult to obtain. To overcome these obstacles, we explore personalized table-to-text generation under a zero-shot setting, by assuming no well-aligned persona-table-text triples are required during training. To this end, we firstly collect a set of unpaired persona information and then propose a semi-supervised approach with contrastive persona distillation (S2P-CPD) to generate personalized context. Specifically, tabular data and persona information are firstly represented as latent variables separately. Then, we devise a latent space fusion technique to distill persona information into the table representation. Besides, a contrastive-based discriminator is employed to guarantee the style consistency between the generated context and its corresponding persona. Experimental results on two benchmarks demonstrate S2P-CPD's ability on keeping both content fidelity and personalized expressions.

Yifu Liu, Xiaoxia Li, Zhiling Luo et al.

Weakly-supervised temporal action localization aims to localize action instances in untrimmed videos with only video-level supervision. We witness that different actions record common phases, e.g., the run-up in the HighJump and LongJump. These different actions are defined as conjoint actions, whose rest parts are definite phases, e.g., leaping over the bar in a HighJump. Compared with the common phases, the definite phases are more easily localized in existing researches. Most of them formulate this task as a Multiple Instance Learning paradigm, in which the common phases are tended to be confused with the background, and affect the localization completeness of the conjoint actions. To tackle this challenge, we propose a Joint of Common and Definite phases Network (JCDNet) by improving feature discriminability of the conjoint actions. Specifically, we design a Class-Aware Discriminative module to enhance the contribution of the common phases in classification by the guidance of the coarse definite-phase features. Besides, we introduce a temporal attention module to learn robust action-ness scores via modeling temporal dependencies, distinguishing the common phases from the background. Extensive experiments on three datasets (THUMOS14, ActivityNetv1.2, and a conjoint-action subset) demonstrate that JCDNet achieves competitive performance against the state-of-the-art methods. Keywords: weakly-supervised learning, temporal action localization, conjoint action

Wei Zhou, Zhou Wang

Omnidirectional image quality assessment (OIQA) aims to predict the perceptual quality of omnidirectional images that cover the whole 180$\times$360$^{\circ}$ viewing range of the visual environment. Here we propose a blind/no-reference OIQA method named S$^2$ that bridges the gap between low-level statistics and high-level semantics of omnidirectional images. Specifically, statistic and semantic features are extracted in separate paths from multiple local viewports and the hallucinated global omnidirectional image, respectively. A quality regression along with a weighting process is then followed that maps the extracted quality-aware features to a perceptual quality prediction. Experimental results demonstrate that the proposed S$^2$ method offers highly competitive performance against state-of-the-art methods.

Minxuan Lv, Chengwei Dai, Kun Li et al.

Neural network models are vulnerable to adversarial examples, and adversarial transferability further increases the risk of adversarial attacks. Current methods based on transferability often rely on substitute models, which can be impractical and costly in real-world scenarios due to the unavailability of training data and the victim model's structural details. In this paper, we propose a novel approach that directly constructs adversarial examples by extracting transferable features across various tasks. Our key insight is that adversarial transferability can extend across different tasks. Specifically, we train a sequence-to-sequence generative model named CT-GAT using adversarial sample data collected from multiple tasks to acquire universal adversarial features and generate adversarial examples for different tasks. We conduct experiments on ten distinct datasets, and the results demonstrate that our method achieves superior attack performance with small cost.

Jingjing Xu, Wei Zhou, Zijian Yang et al.

Varying-size models are often required to deploy ASR systems under different hardware and/or application constraints such as memory and latency. To avoid redundant training and optimization efforts for individual models of different sizes, we present the dynamic encoder size approach, which jointly trains multiple performant models within one supernet from scratch. These subnets of various sizes are layer-wise pruned from the supernet, and thus, enjoy full parameter sharing. By combining score-based pruning with supernet training, we propose two novel methods, Simple-Top-k and Iterative-Zero-Out, to automatically select the best-performing subnets in a data-driven manner, avoiding resource-intensive search efforts. Our experiments using CTC on both Librispeech and TED-LIUM-v2 corpora show that our methods can achieve on-par performance as individually trained models of each size category. Also, our approach consistently brings small performance improvements for the full-size supernet.

Zijian Yang, Wei Zhou, Ralf Schlüter et al.

In this work, we investigate the effect of language models (LMs) with different context lengths and label units (phoneme vs. word) used in sequence discriminative training for phoneme-based neural transducers. Both lattice-free and N-best-list approaches are examined. For lattice-free methods with phoneme-level LMs, we propose a method to approximate the context history to employ LMs with full-context dependency. This approximation can be extended to arbitrary context length and enables the usage of word-level LMs in lattice-free methods. Moreover, a systematic comparison is conducted across lattice-free and N-best-list-based methods. Experimental results on Librispeech show that using the word-level LM in training outperforms the phoneme-level LM. Besides, we find that the context size of the LM used for probability computation has a limited effect on performance. Moreover, our results reveal the pivotal importance of the hypothesis space quality in sequence discriminative training.

Rohan Agarwal, Wei Zhou, Xiaofeng Wu et al.

Reconstructing a 3D object from a 2D image is a well-researched vision problem, with many kinds of deep learning techniques having been tried. Most commonly, 3D convolutional approaches are used, though previous work has shown state-of-the-art methods using 2D convolutions that are also significantly more efficient to train. With the recent rise of transformers for vision tasks, often outperforming convolutional methods, along with some earlier attempts to use transformers for 3D object reconstruction, we set out to use visual transformers in place of convolutions in existing efficient, high-performing techniques for 3D object reconstruction in order to achieve superior results on the task. Using a transformer-based encoder and decoder to predict 3D structure from 2D images, we achieve accuracy similar or superior to the baseline approach. This study serves as evidence for the potential of visual transformers in the task of 3D object reconstruction.

Yao Hu, Guohua Geng, Kang Li et al.

The repairing work of terracotta warriors in Emperor Qinshihuang Mausoleum Site Museum is handcrafted by experts, and the increasing amounts of unearthed pieces of terracotta warriors make the archaeologists too challenging to conduct the restoration of terracotta warriors efficiently. We hope to segment the 3D point cloud data of the terracotta warriors automatically and store the fragment data in the database to assist the archaeologists in matching the actual fragments with the ones in the database, which could result in higher repairing efficiency of terracotta warriors. Moreover, the existing 3D neural network research is mainly focusing on supervised classification, clustering, unsupervised representation, and reconstruction. There are few pieces of researches concentrating on unsupervised point cloud part segmentation. In this paper, we present SRG-Net for 3D point clouds of terracotta warriors to address these problems. Firstly, we adopt a customized seed-region-growing algorithm to segment the point cloud coarsely. Then we present a supervised segmentation and unsupervised reconstruction networks to learn the characteristics of 3D point clouds. Finally, we combine the SRG algorithm with our improved CNN(convolution neural network) using a refinement method. This pipeline is called SRG-Net, which aims at conducting segmentation tasks on the terracotta warriors. Our proposed SRG-Net is evaluated on the terracotta warrior data and ShapeNet dataset by measuring the accuracy and the latency. The experimental results show that our SRG-Net outperforms the state-of-the-art methods. Our code is available at https://github.com/hyoau/SRG-Net.

Wei Zhou, Eugen Beck, Simon Berger et al.

Modern public ASR tools usually provide rich support for training various sequence-to-sequence (S2S) models, but rather simple support for decoding open-vocabulary scenarios only. For closed-vocabulary scenarios, public tools supporting lexical-constrained decoding are usually only for classical ASR, or do not support all S2S models. To eliminate this restriction on research possibilities such as modeling unit choice, we present RASR2 in this work, a research-oriented generic S2S decoder implemented in C++. It offers a strong flexibility/compatibility for various S2S models, language models, label units/topologies and neural network architectures. It provides efficient decoding for both open- and closed-vocabulary scenarios based on a generalized search framework with rich support for different search modes and settings. We evaluate RASR2 with a wide range of experiments on both switchboard and Librispeech corpora. Our source code is public online.

Wei Zhou, Qian Wang, Weiwei Jin et al.

Recently, graph-based and Transformer-based deep learning networks have demonstrated excellent performances on various point cloud tasks. Most of the existing graph methods are based on static graph, which take a fixed input to establish graph relations. Moreover, many graph methods apply maximization and averaging to aggregate neighboring features, so that only a single neighboring point affects the feature of centroid or different neighboring points have the same influence on the centroid's feature, which ignoring the correlation and difference between points. Most Transformer-based methods extract point cloud features based on global attention and lack the feature learning on local neighbors. To solve the problems of these two types of models, we propose a new feature extraction block named Graph Transformer and construct a 3D point point cloud learning network called GTNet to learn features of point clouds on local and global patterns. Graph Transformer integrates the advantages of graph-based and Transformer-based methods, and consists of Local Transformer and Global Transformer modules. Local Transformer uses a dynamic graph to calculate all neighboring point weights by intra-domain cross-attention with dynamically updated graph relations, so that every neighboring point could affect the features of centroid with different weights; Global Transformer enlarges the receptive field of Local Transformer by a global self-attention. In addition, to avoid the disappearance of the gradient caused by the increasing depth of network, we conduct residual connection for centroid features in GTNet; we also adopt the features of centroid and neighbors to generate the local geometric descriptors in Local Transformer to strengthen the local information learning capability of the model. Finally, we use GTNet for shape classification, part segmentation and semantic segmentation tasks in this paper.

Liqiang Jing, Xuemeng Song, Xuming Lin et al.

Existing data-to-text generation efforts mainly focus on generating a coherent text from non-linguistic input data, such as tables and attribute-value pairs, but overlook that different application scenarios may require texts of different styles. Inspired by this, we define a new task, namely stylized data-to-text generation, whose aim is to generate coherent text for the given non-linguistic data according to a specific style. This task is non-trivial, due to three challenges: the logic of the generated text, unstructured style reference, and biased training samples. To address these challenges, we propose a novel stylized data-to-text generation model, named StyleD2T, comprising three components: logic planning-enhanced data embedding, mask-based style embedding, and unbiased stylized text generation. In the first component, we introduce a graph-guided logic planner for attribute organization to ensure the logic of generated text. In the second component, we devise feature-level mask-based style embedding to extract the essential style signal from the given unstructured style reference. In the last one, pseudo triplet augmentation is utilized to achieve unbiased text generation, and a multi-condition based confidence assignment function is designed to ensure the quality of pseudo samples. Extensive experiments on a newly collected dataset from Taobao have been conducted, and the results show the superiority of our model over existing methods.

Guangyan Zhang, Kaitao Song, Xu Tan et al.

Recently, leveraging BERT pre-training to improve the phoneme encoder in text to speech (TTS) has drawn increasing attention. However, the works apply pre-training with character-based units to enhance the TTS phoneme encoder, which is inconsistent with the TTS fine-tuning that takes phonemes as input. Pre-training only with phonemes as input can alleviate the input mismatch but lack the ability to model rich representations and semantic information due to limited phoneme vocabulary. In this paper, we propose MixedPhoneme BERT, a novel variant of the BERT model that uses mixed phoneme and sup-phoneme representations to enhance the learning capability. Specifically, we merge the adjacent phonemes into sup-phonemes and combine the phoneme sequence and the merged sup-phoneme sequence as the model input, which can enhance the model capacity to learn rich contextual representations. Experiment results demonstrate that our proposed Mixed-Phoneme BERT significantly improves the TTS performance with 0.30 CMOS gain compared with the FastSpeech 2 baseline. The Mixed-Phoneme BERT achieves 3x inference speedup and similar voice quality to the previous TTS pre-trained model PnG BERT

Wei Zhou, Xiongkuo Min, Hong Li et al.

Quality of experience (QoE) assessment for adaptive video streaming plays a significant role in advanced network management systems. It is especially challenging in case of dynamic adaptive streaming schemes over HTTP (DASH) which has increasingly complex characteristics including additional playback issues. In this paper, we provide a brief overview of adaptive video streaming quality assessment. Upon our review of related works, we analyze and compare different variations of objective QoE assessment models with or without using machine learning techniques for adaptive video streaming. Through the performance analysis, we observe that hybrid models perform better than both quality-of-service (QoS) driven QoE approaches and signal fidelity measurement. Moreover, the machine learning-based model slightly outperforms the model without using machine learning for the same setting. In addition, we find that existing video streaming QoE assessment models still have limited performance, which makes it difficult to be applied in practical communication systems. Therefore, based on the success of deep learned feature representations for traditional video quality prediction, we also apply the off-the-shelf deep convolutional neural network (DCNN) to evaluate the perceptual quality of streaming videos, where the spatio-temporal properties of streaming videos are taken into consideration. Experiments demonstrate its superiority, which sheds light on the future development of specifically designed deep learning frameworks for adaptive video streaming quality assessment. We believe this survey can serve as a guideline for QoE assessment of adaptive video streaming.

Yu Qiao, Sourabh Zanwar, Rishab Bhattacharyya et al.

One of the key communicative competencies is the ability to maintain fluency in monologic speech and the ability to produce sophisticated language to argue a position convincingly. In this paper we aim to predict TED talk-style affective ratings in a crowdsourced dataset of argumentative speech consisting of 7 hours of speech from 110 individuals. The speech samples were elicited through task prompts relating to three debating topics. The samples received a total of 2211 ratings from 737 human raters pertaining to 14 affective categories. We present an effective approach to the classification task of predicting these categories through fine-tuning a model pre-trained on a large dataset of TED talks public speeches. We use a combination of fluency features derived from a state-of-the-art automatic speech recognition system and a large set of human-interpretable linguistic features obtained from an automatic text analysis system. Classification accuracy was greater than 60% for all 14 rating categories, with a peak performance of 72% for the rating category 'informative'. In a secondary experiment, we determined the relative importance of features from different groups using SP-LIME.

Wei Zhou, Zuoyun Zheng, Ralf Schlüter et al.

The mismatch between an external language model (LM) and the implicitly learned internal LM (ILM) of RNN-Transducer (RNN-T) can limit the performance of LM integration such as simple shallow fusion. A Bayesian interpretation suggests to remove this sequence prior as ILM correction. In this work, we study various ILM correction-based LM integration methods formulated in a common RNN-T framework. We provide a decoding interpretation on two major reasons for performance improvement with ILM correction, which is further experimentally verified with detailed analysis. We also propose an exact-ILM training framework by extending the proof given in the hybrid autoregressive transducer, which enables a theoretical justification for other ILM approaches. Systematic comparison is conducted for both in-domain and cross-domain evaluation on the Librispeech and TED-LIUM Release 2 corpora, respectively. Our proposed exact-ILM training can further improve the best ILM method.

Binlong Zhang, Wei Zhou

Chinese sentiment analysis (CSA) has always been one of the challenges in natural language processing due to its complexity and uncertainty. Transformer has succeeded in capturing semantic features, but it uses position encoding to capture sequence features, which has great shortcomings compared with the recurrent model. In this paper, we propose T-E-GRU for Chinese sentiment analysis, which combine transformer encoder and GRU. We conducted experiments on three Chinese comment datasets. In view of the confusion of punctuation marks in Chinese comment texts, we selectively retain some punctuation marks with sentence segmentation ability. The experimental results show that T-E-GRU outperforms classic recurrent model and recurrent model with attention.

Yao Hu, Guohua Geng, Kang Li et al.

The repairing work of terracotta warriors in Emperor Qinshihuang Mausoleum Site Museum is handcrafted by experts, and the increasing amounts of unearthed pieces of terracotta warriors make the archaeologists too challenging to conduct the restoration of terracotta warriors efficiently. We hope to segment the 3D point cloud data of the terracotta warriors automatically and store the fragment data in the database to assist the archaeologists in matching the actual fragments with the ones in the database, which could result in higher repairing efficiency of terracotta warriors. Moreover, the existing 3D neural network research is mainly focusing on supervised classification, clustering, unsupervised representation, and reconstruction. There are few pieces of researches concentrating on unsupervised point cloud part segmentation. In this paper, we present SRG-Net for 3D point clouds of terracotta warriors to address these problems. Firstly, we adopt a customized seed-region-growing algorithm to segment the point cloud coarsely. Then we present a supervised segmentation and unsupervised reconstruction networks to learn the characteristics of 3D point clouds. Finally, we combine the SRG algorithm with our improved CNN using a refinement method. This pipeline is called SRG-Net, which aims at conducting segmentation tasks on the terracotta warriors. Our proposed SRG-Net is evaluated on the terracotta warriors data and ShapeNet dataset by measuring the accuracy and the latency. The experimental results show that our SRG-Net outperforms the state-of-the-art methods. Our code is shown in Code File 1~\cite{Srgnet_2021}.

Jun Fu, Chen Hou, Wei Zhou et al.

In no-reference 360-degree image quality assessment (NR 360IQA), graph convolutional networks (GCNs), which model interactions between viewports through graphs, have achieved impressive performance. However, prevailing GCN-based NR 360IQA methods suffer from three main limitations. First, they only use high-level features of the distorted image to regress the quality score, while the human visual system (HVS) scores the image based on hierarchical features. Second, they simplify complex high-order interactions between viewports in a pairwise fashion through graphs. Third, in the graph construction, they only consider spatial locations of viewports, ignoring its content characteristics. Accordingly, to address these issues, we propose an adaptive hypergraph convolutional network for NR 360IQA, denoted as AHGCN. Specifically, we first design a multi-level viewport descriptor for extracting hierarchical representations from viewports. Then, we model interactions between viewports through hypergraphs, where each hyperedge connects two or more viewports. In the hypergraph construction, we build a location-based hyperedge and a content-based hyperedge for each viewport. Experimental results on two public 360IQA databases demonstrate that our proposed approach has a clear advantage over state-of-the-art full-reference and no-reference IQA models.

Chunyuan Yuan, Wanhui Qian, Qianwen Ma et al.

The rapid development of social media changes the lifestyle of people and simultaneously provides an ideal place for publishing and disseminating rumors, which severely exacerbates social panic and triggers a crisis of social trust. Early content-based methods focused on finding clues from the text and user profiles for rumor detection. Recent studies combine the stances of users' comments with news content to capture the difference between true and false rumors. Although the user's stance is effective for rumor detection, the manual labeling process is time-consuming and labor-intensive, which limits the application of utilizing it to facilitate rumor detection. In this paper, we first finetune a pre-trained BERT model on a small labeled dataset and leverage this model to annotate weak stance labels for users' comment data to overcome the problem mentioned above. Then, we propose a novel Stance-aware Reinforcement Learning Framework (SRLF) to select high-quality labeled stance data for model training and rumor detection. Both the stance selection and rumor detection tasks are optimized simultaneously to promote both tasks mutually. We conduct experiments on two commonly used real-world datasets. The experimental results demonstrate that our framework outperforms the state-of-the-art models significantly, which confirms the effectiveness of the proposed framework.

Jianzhao Liu, Wei Zhou, Jiahua Xu et al.

Existing blind image quality assessment (BIQA) methods are mostly designed in a disposable way and cannot evolve with unseen distortions adaptively, which greatly limits the deployment and application of BIQA models in real-world scenarios. To address this problem, we propose a novel Lifelong blind Image Quality Assessment (LIQA) approach, targeting to achieve the lifelong learning of BIQA. Without accessing to previous training data, our proposed LIQA can not only learn new distortions, but also mitigate the catastrophic forgetting of seen distortions. Specifically, we adopt the Split-and-Merge distillation strategy to train a single-head network that makes task-agnostic predictions. In the split stage, we first employ a distortion-specific generator to obtain the pseudo features of each seen distortion. Then, we use an auxiliary multi-head regression network to generate the predicted quality of each seen distortion. In the merge stage, we replay the pseudo features paired with pseudo labels to distill the knowledge of multiple heads, which can build the final regressed single head. Experimental results demonstrate that the proposed LIQA method can handle the continuous shifts of different distortion types and even datasets. More importantly, our LIQA model can achieve stable performance even if the task sequence is long.

Wei Zhou, Mohammad Zeineldeen, Zuoyun Zheng et al.

Subword units are commonly used for end-to-end automatic speech recognition (ASR), while a fully acoustic-oriented subword modeling approach is somewhat missing. We propose an acoustic data-driven subword modeling (ADSM) approach that adapts the advantages of several text-based and acoustic-based subword methods into one pipeline. With a fully acoustic-oriented label design and learning process, ADSM produces acoustic-structured subword units and acoustic-matched target sequence for further ASR training. The obtained ADSM labels are evaluated with different end-to-end ASR approaches including CTC, RNN-Transducer and attention models. Experiments on the LibriSpeech corpus show that ADSM clearly outperforms both byte pair encoding (BPE) and pronunciation-assisted subword modeling (PASM) in all cases. Detailed analysis shows that ADSM achieves acoustically more logical word segmentation and more balanced sequence length, and thus, is suitable for both time-synchronous and label-synchronous models. We also briefly describe how to apply acoustic-based subword regularization and unseen text segmentation using ADSM.

Yu Qiao, Wei Zhou, Elma Kerz et al.

In recent years, automated approaches to assessing linguistic complexity in second language (L2) writing have made significant progress in gauging learner performance, predicting human ratings of the quality of learner productions, and benchmarking L2 development. In contrast, there is comparatively little work in the area of speaking, particularly with respect to fully automated approaches to assessing L2 spontaneous speech. While the importance of a well-performing ASR system is widely recognized, little research has been conducted to investigate the impact of its performance on subsequent automatic text analysis. In this paper, we focus on this issue and examine the impact of using a state-of-the-art ASR system for subsequent automatic analysis of linguistic complexity in spontaneously produced L2 speech. A set of 30 selected measures were considered, falling into four categories: syntactic, lexical, n-gram frequency, and information-theoretic measures. The agreement between the scores for these measures obtained on the basis of ASR-generated vs. manual transcriptions was determined through correlation analysis. A more differential effect of ASR performance on specific types of complexity measures when controlling for task type effects is also presented.

Wei Zhou, Albert Zeyer, André Merboldt et al.

With the advent of direct models in automatic speech recognition (ASR), the formerly prevalent frame-wise acoustic modeling based on hidden Markov models (HMM) diversified into a number of modeling architectures like encoder-decoder attention models, transducer models and segmental models (direct HMM). While transducer models stay with a frame-level model definition, segmental models are defined on the level of label segments directly. While (soft-)attention-based models avoid explicit alignment, transducer and segmental approach internally do model alignment, either by segment hypotheses or, more implicitly, by emitting so-called blank symbols. In this work, we prove that the widely used class of RNN-Transducer models and segmental models (direct HMM) are equivalent and therefore show equal modeling power. It is shown that blank probabilities translate into segment length probabilities and vice versa. In addition, we provide initial experiments investigating decoding and beam-pruning, comparing time-synchronous and label-/segment-synchronous search strategies and their properties using the same underlying model.

Simeng Sun, Tao Yu, Jiahua Xu et al.

A good distortion representation is crucial for the success of deep blind image quality assessment (BIQA). However, most previous methods do not effectively model the relationship between distortions or the distribution of samples with the same distortion type but different distortion levels. In this work, we start from the analysis of the relationship between perceptual image quality and distortion-related factors, such as distortion types and levels. Then, we propose a Distortion Graph Representation (DGR) learning framework for IQA, named GraphIQA, in which each distortion is represented as a graph, i.e., DGR. One can distinguish distortion types by learning the contrast relationship between these different DGRs, and infer the ranking distribution of samples from different levels in a DGR. Specifically, we develop two sub-networks to learn the DGRs: a) Type Discrimination Network (TDN) that aims to embed DGR into a compact code for better discriminating distortion types and learning the relationship between types; b) Fuzzy Prediction Network (FPN) that aims to extract the distributional characteristics of the samples in a DGR and predicts fuzzy degrees based on a Gaussian prior. Experiments show that our GraphIQA achieves the state-of-the-art performance on many benchmark datasets of both synthetic and authentic distortions.

Yiying Li, Wei Zhou, Huaimin Wang et al.

Federated learning (FL) enables distributed participants to collectively learn a strong global model without sacrificing their individual data privacy. Mainstream FL approaches require each participant to share a common network architecture and further assume that data are are sampled IID across participants. However, in real-world deployments participants may require heterogeneous network architectures; and the data distribution is almost certainly non-uniform across participants. To address these issues we introduce FedH2L, which is agnostic to both the model architecture and robust to different data distributions across participants. In contrast to approaches sharing parameters or gradients, FedH2L relies on mutual distillation, exchanging only posteriors on a shared seed set between participants in a decentralized manner. This makes it extremely bandwidth efficient, model agnostic, and crucially produces models capable of performing well on the whole data distribution when learning from heterogeneous silos.

Wei Zhou, Simon Berger, Ralf Schlüter et al.

To join the advantages of classical and end-to-end approaches for speech recognition, we present a simple, novel and competitive approach for phoneme-based neural transducer modeling. Different alignment label topologies are compared and word-end-based phoneme label augmentation is proposed to improve performance. Utilizing the local dependency of phonemes, we adopt a simplified neural network structure and a straightforward integration with the external word-level language model to preserve the consistency of seq-to-seq modeling. We also present a simple, stable and efficient training procedure using frame-wise cross-entropy loss. A phonetic context size of one is shown to be sufficient for the best performance. A simplified scheduled sampling approach is applied for further improvement and different decoding approaches are briefly compared. The overall performance of our best model is comparable to state-of-the-art (SOTA) results for the TED-LIUM Release 2 and Switchboard corpora.

Mohammad Zeineldeen, Albert Zeyer, Wei Zhou et al.

Following the rationale of end-to-end modeling, CTC, RNN-T or encoder-decoder-attention models for automatic speech recognition (ASR) use graphemes or grapheme-based subword units based on e.g. byte-pair encoding (BPE). The mapping from pronunciation to spelling is learned completely from data. In contrast to this, classical approaches to ASR employ secondary knowledge sources in the form of phoneme lists to define phonetic output labels and pronunciation lexica. In this work, we do a systematic comparison between grapheme- and phoneme-based output labels for an encoder-decoder-attention ASR model. We investigate the use of single phonemes as well as BPE-based phoneme groups as output labels of our model. To preserve a simplified and efficient decoder design, we also extend the phoneme set by auxiliary units to be able to distinguish homophones. Experiments performed on the Switchboard 300h and LibriSpeech benchmarks show that phoneme-based modeling is competitive to grapheme-based encoder-decoder-attention modeling.

Wei Zhou, Ralf Schlüter, Hermann Ney

In hybrid HMM based speech recognition, LSTM language models have been widely applied and achieved large improvements. The theoretical capability of modeling any unlimited context suggests that no recombination should be applied in decoding. This motivates to reconsider full summation over the HMM-state sequences instead of Viterbi approximation in decoding. We explore the potential gain from more accurate probabilities in terms of decision making and apply the full-sum decoding with a modified prefix-tree search framework. The proposed full-sum decoding is evaluated on both Switchboard and Librispeech corpora. Different models using CE and sMBR training criteria are used. Additionally, both MAP and confusion network decoding as approximated variants of general Bayes decision rule are evaluated. Consistent improvements over strong baselines are achieved in almost all cases without extra cost. We also discuss tuning effort, efficiency and some limitations of full-sum decoding.

Wei Zhou, Wilfried Michel, Kazuki Irie et al.

We present a complete training pipeline to build a state-of-the-art hybrid HMM-based ASR system on the 2nd release of the TED-LIUM corpus. Data augmentation using SpecAugment is successfully applied to improve performance on top of our best SAT model using i-vectors. By investigating the effect of different maskings, we achieve improvements from SpecAugment on hybrid HMM models without increasing model size and training time. A subsequent sMBR training is applied to fine-tune the final acoustic model, and both LSTM and Transformer language models are trained and evaluated. Our best system achieves a 5.6% WER on the test set, which outperforms the previous state-of-the-art by 27% relative.

Zhenxin Fu, Feng Ji, Wenpeng Hu et al.

Information-seeking conversation system aims at satisfying the information needs of users through conversations. Text matching between a user query and a pre-collected question is an important part of the information-seeking conversation in E-commerce. In the practical scenario, a sort of questions always correspond to a same answer. Naturally, these questions can form a bag. Learning the matching between user query and bag directly may improve the conversation performance, denoted as query-bag matching. Inspired by such opinion, we propose a query-bag matching model which mainly utilizes the mutual coverage between query and bag and measures the degree of the content in the query mentioned by the bag, and vice verse. In addition, the learned bag representation in word level helps find the main points of a bag in a fine grade and promotes the query-bag matching performance. Experiments on two datasets show the effectiveness of our model.

Eugen Beck, Wei Zhou, Ralf Schlüter et al.

LSTM based language models are an important part of modern LVCSR systems as they significantly improve performance over traditional backoff language models. Incorporating them efficiently into decoding has been notoriously difficult. In this paper we present an approach based on a combination of one-pass decoding and lattice rescoring. We perform decoding with the LSTM-LM in the first pass but recombine hypothesis that share the last two words, afterwards we rescore the resulting lattice. We run our systems on GPGPU equipped machines and are able to produce competitive results on the Hub5'00 and Librispeech evaluation corpora with a runtime better than real-time. In addition we shortly investigate the possibility to carry out the full sum over all state-sequences belonging to a given word-hypothesis during decoding without recombination.