arXiv subjects

Yu Wu

Publications and source records attributed to Yu Wu.

At least 217 records · Page 12Linked to original sources

arXiv2021

Contrastive Video-Language Segmentation

We focus on the problem of segmenting a certain object referred by a natural language sentence in video content, at the core of formulating a pinpoint vision-language relation. While existing attempts mainly construct such relation in an implicit way, i.e., grid-level multi-modal feature fusion, it has been proven problematic to distinguish semantically similar objects under this paradigm. In this work, we propose to interwind the visual and linguistic modalities in an explicit way via the contrastive learning objective, which directly aligns the referred object and the language description and separates the unreferred content apart across frames. Moreover, to remedy for the degradation problem, we present two complementary hard instance mining strategies, i.e., Language-relevant Channel Filter and Relative Hard Instance Construction. They encourage the network to exclude visual-distinguishable feature and to focus on easy-confused objects during the contrastive training. Extensive experiments on two benchmarks, i.e., A2D Sentences and J-HMDB Sentences, quantitatively demonstrate the state-of-the-arts performance of our method and qualitatively show the more accurate distinguishment between semantically similar objects over baselines.

cs.CV↗

arXiv2021

Knowledge Enhanced Fine-Tuning for Better Handling Unseen Entities in Dialogue Generation

Although pre-training models have achieved great success in dialogue generation, their performance drops dramatically when the input contains an entity that does not appear in pre-training and fine-tuning datasets (unseen entity). To address this issue, existing methods leverage an external knowledge base to generate appropriate responses. In real-world scenario, the entity may not be included by the knowledge base or suffer from the precision of knowledge retrieval. To deal with this problem, instead of introducing knowledge base as the input, we force the model to learn a better semantic representation by predicting the information in the knowledge base, only based on the input context. Specifically, with the help of a knowledge base, we introduce two auxiliary training objectives: 1) Interpret Masked Word, which conjectures the meaning of the masked entity given the context; 2) Hypernym Generation, which predicts the hypernym of the entity based on the context. Experiment results on two dialogue corpus verify the effectiveness of our methods under both knowledge available and unavailable settings.

cs.CL↗

arXiv2021

Detecting Speaker Personas from Conversational Texts

Personas are useful for dialogue response prediction. However, the personas used in current studies are pre-defined and hard to obtain before a conversation. To tackle this issue, we study a new task, named Speaker Persona Detection (SPD), which aims to detect speaker personas based on the plain conversational text. In this task, a best-matched persona is searched out from candidates given the conversational text. This is a many-to-many semantic matching task because both contexts and personas in SPD are composed of multiple sentences. The long-term dependency and the dynamic redundancy among these sentences increase the difficulty of this task. We build a dataset for SPD, dubbed as Persona Match on Persona-Chat (PMPC). Furthermore, we evaluate several baseline models and propose utterance-to-profile (U2P) matching networks for this task. The U2P models operate at a fine granularity which treat both contexts and personas as sets of multiple sequences. Then, each sequence pair is scored and an interpretable overall score is obtained for a context-persona pair through aggregation. Evaluation results show that the U2P models outperform their baseline counterparts significantly.

cs.CL↗

arXiv2021

Flexible Clustered Federated Learning for Client-Level Data Distribution Shift

Federated Learning (FL) enables the multiple participating devices to collaboratively contribute to a global neural network model while keeping the training data locally. Unlike the centralized training setting, the non-IID, imbalanced (statistical heterogeneity) and distribution shifted training data of FL is distributed in the federated network, which will increase the divergences between the local models and the global model, further degrading performance. In this paper, we propose a flexible clustered federated learning (CFL) framework named FlexCFL, in which we 1) group the training of clients based on the similarities between the clients' optimization directions for lower training divergence; 2) implement an efficient newcomer device cold start mechanism for framework scalability and practicality; 3) flexibly migrate clients to meet the challenge of client-level data distribution shift. FlexCFL can achieve improvements by dividing joint optimization into groups of sub-optimization and can strike a balance between accuracy and communication efficiency in the distribution shift environment. The convergence and complexity are analyzed to demonstrate the efficiency of FlexCFL. We also evaluate FlexCFL on several open datasets and made comparisons with related CFL frameworks. The results show that FlexCFL can significantly improve absolute test accuracy by +10.6% on FEMNIST compared to FedAvg, +3.5% on FashionMNIST compared to FedProx, +8.4% on MNIST compared to FeSEM. The experiment results show that FlexCFL is also communication efficient in the distribution shift environment.

cs.LG↗

arXiv2021

UniSpeech at scale: An Empirical Study of Pre-training Method on Large-Scale Speech Recognition Dataset

Recently, there has been a vast interest in self-supervised learning (SSL) where the model is pre-trained on large scale unlabeled data and then fine-tuned on a small labeled dataset. The common wisdom is that SSL helps resource-limited tasks in which only a limited amount of labeled data is available. The benefit of SSL keeps diminishing when the labeled training data amount increases. To our best knowledge, at most a few thousand hours of labeled data was used in the study of SSL. In contrast, the industry usually uses tens of thousands of hours of labeled data to build high-accuracy speech recognition (ASR) systems for resource-rich languages. In this study, we take the challenge to investigate whether and how SSL can improve the ASR accuracy of a state-of-the-art production-scale Transformer-Transducer model, which was built with 65 thousand hours of anonymized labeled EN-US data.

eess.AS↗

arXiv2021

Investigation of Practical Aspects of Single Channel Speech Separation for ASR

Speech separation has been successfully applied as a frontend processing module of conversation transcription systems thanks to its ability to handle overlapped speech and its flexibility to combine with downstream tasks such as automatic speech recognition (ASR). However, a speech separation model often introduces target speech distortion, resulting in a sub-optimum word error rate (WER). In this paper, we describe our efforts to improve the performance of a single channel speech separation system. Specifically, we investigate a two-stage training scheme that firstly applies a feature level optimization criterion for pretraining, followed by an ASR-oriented optimization criterion using an end-to-end (E2E) speech recognition model. Meanwhile, to keep the model light-weight, we introduce a modified teacher-student learning technique for model compression. By combining those approaches, we achieve a absolute average WER improvement of 2.70% and 0.77% using models with less than 10M parameters compared with the previous state-of-the-art results on the LibriCSS dataset for utterance-wise evaluation and continuous evaluation, respectively

eess.AS↗

arXiv2021

Saying the Unseen: Video Descriptions via Dialog Agents

Current vision and language tasks usually take complete visual data (e.g., raw images or videos) as input, however, practical scenarios may often consist the situations where part of the visual information becomes inaccessible due to various reasons e.g., restricted view with fixed camera or intentional vision block for security concerns. As a step towards the more practical application scenarios, we introduce a novel task that aims to describe a video using the natural language dialog between two agents as a supplementary information source given incomplete visual data. Different from most existing vision-language tasks where AI systems have full access to images or video clips, which may reveal sensitive information such as recognizable human faces or voices, we intentionally limit the visual input for AI systems and seek a more secure and transparent information medium, i.e., the natural language dialog, to supplement the missing visual information. Specifically, one of the intelligent agents - Q-BOT - is given two semantic segmented frames from the beginning and the end of the video, as well as a finite number of opportunities to ask relevant natural language questions before describing the unseen video. A-BOT, the other agent who has access to the entire video, assists Q-BOT to accomplish the goal by answering the asked questions. We introduce two different experimental settings with either a generative (i.e., agents generate questions and answers freely) or a discriminative (i.e., agents select the questions and answers from candidates) internal dialog generation process. With the proposed unified QA-Cooperative networks, we experimentally demonstrate the knowledge transfer process between the two dialog agents and the effectiveness of using the natural language dialog as a supplement for incomplete implicit visions.

cs.CV↗

arXiv2021

On Commonsense Cues in BERT for Solving Commonsense Tasks

BERT has been used for solving commonsense tasks such as CommonsenseQA. While prior research has found that BERT does contain commonsense information to some extent, there has been work showing that pre-trained models can rely on spurious associations (e.g., data bias) rather than key cues in solving sentiment classification and other problems. We quantitatively investigate the presence of structural commonsense cues in BERT when solving commonsense tasks, and the importance of such cues for the model prediction. Using two different measures, we find that BERT does use relevant knowledge for solving the task, and the presence of commonsense knowledge is positively correlated to the model accuracy.

cs.CL↗

arXiv2021

UniSpeech: Unified Speech Representation Learning with Labeled and Unlabeled Data

In this paper, we propose a unified pre-training approach called UniSpeech to learn speech representations with both unlabeled and labeled data, in which supervised phonetic CTC learning and phonetically-aware contrastive self-supervised learning are conducted in a multi-task learning manner. The resultant representations can capture information more correlated with phonetic structures and improve the generalization across languages and domains. We evaluate the effectiveness of UniSpeech for cross-lingual representation learning on public CommonVoice corpus. The results show that UniSpeech outperforms self-supervised pretraining and supervised transfer learning for speech recognition by a maximum of 13.4% and 17.8% relative phone error rate reductions respectively (averaged over all testing languages). The transferability of UniSpeech is also demonstrated on a domain-shift speech recognition task, i.e., a relative word error rate reduction of 6% against the previous approach.

cs.CL↗

arXiv2021

Minimum Word Error Rate Training with Language Model Fusion for End-to-End Speech Recognition

Integrating external language models (LMs) into end-to-end (E2E) models remains a challenging task for domain-adaptive speech recognition. Recently, internal language model estimation (ILME)-based LM fusion has shown significant word error rate (WER) reduction from Shallow Fusion by subtracting a weighted internal LM score from an interpolation of E2E model and external LM scores during beam search. However, on different test sets, the optimal LM interpolation weights vary over a wide range and have to be tuned extensively on well-matched validation sets. In this work, we perform LM fusion in the minimum WER (MWER) training of an E2E model to obviate the need for LM weights tuning during inference. Besides MWER training with Shallow Fusion (MWER-SF), we propose a novel MWER training with ILME (MWER-ILME) where the ILME-based fusion is conducted to generate N-best hypotheses and their posteriors. Additional gradient is induced when internal LM is engaged in MWER-ILME loss computation. During inference, LM weights pre-determined in MWER training enable robust LM integrations on test sets from different domains. Experimented with 30K-hour trained transformer transducers, MWER-ILME achieves on average 8.8% and 5.8% relative WER reductions from MWER and MWER-SF training, respectively, on 6 different test sets

eess.AS↗

arXiv2021

Template-Based Named Entity Recognition Using BART

There is a recent interest in investigating few-shot NER, where the low-resource target domain has different label sets compared with a resource-rich source domain. Existing methods use a similarity-based metric. However, they cannot make full use of knowledge transfer in NER model parameters. To address the issue, we propose a template-based method for NER, treating NER as a language model ranking problem in a sequence-to-sequence framework, where original sentences and statement templates filled by candidate named entity span are regarded as the source sequence and the target sequence, respectively. For inference, the model is required to classify each candidate span based on the corresponding template scores. Our experiments demonstrate that the proposed method achieves 92.55% F1 score on the CoNLL03 (rich-resource task), and significantly better than fine-tuning BERT 10.88%, 15.34%, and 11.73% F1 score on the MIT Movie, the MIT Restaurant, and the ATIS (low-resource task), respectively.

cs.CL↗

arXiv2021

Large-Scale Pre-Training of End-to-End Multi-Talker ASR for Meeting Transcription with Single Distant Microphone

Transcribing meetings containing overlapped speech with only a single distant microphone (SDM) has been one of the most challenging problems for automatic speech recognition (ASR). While various approaches have been proposed, all previous studies on the monaural overlapped speech recognition problem were based on either simulation data or small-scale real data. In this paper, we extensively investigate a two-step approach where we first pre-train a serialized output training (SOT)-based multi-talker ASR by using large-scale simulation data and then fine-tune the model with a small amount of real meeting data. Experiments are conducted by utilizing 75 thousand (K) hours of our internal single-talker recording to simulate a total of 900K hours of multi-talker audio segments for supervised pre-training. With fine-tuning on the 70 hours of the AMI-SDM training data, our SOT ASR model achieves a word error rate (WER) of 21.2% for the AMI-SDM evaluation set while automatically counting speakers in each test segment. This result is not only significantly better than the previous state-of-the-art WER of 36.4% with oracle utterance boundary information but also better than a result by a similarly fine-tuned single-talker ASR model applied to beamformed audio.

eess.AS↗

arXiv2021

Dissipative Transport and Phonon Scattering Suppression via Valley Engineering in Single-Layer Antimonene and Arsenene Field-Effect Transistors

Two-dimensional (2D) semiconductors are promising channel materials for next-generation field-effect transistors (FETs) thanks to their unique mechanical properties and enhanced electrostatic control. However, the performance of these devices can be strongly limited by the scattering processes between carriers and phonons, usually occurring at high rates in 2D materials. Here, we use quantum transport simulations calibrated on first-principle computations to report on dissipative transport in antimonene and arsenene $n$-type FETs at the scaling limit. We show that the widely-used approximations of either ballistic transport or simple acoustic deformation potential scattering result in large overestimation of the ON current, due to neglecting the dominant intervalley and optical phonon scattering processes. We additionally investigate valley engineering strategy [Nano Lett. \textbf{19}, 3723 (2019)] to improve the device performance by removing the valley degeneracy and suppressing most of the intervalley scattering channels via an uniaxial strain along the zigzag direction. The method is applicable to other similar 2D semiconductors characterized by multivalley transport.

cond-mat.mes-hall↗

arXiv2021

Virtual Reality Sickness Mitigation Methods: A Comparative Study in a Racing Game

Using virtual reality (VR) head-mounted displays (HMDs) can induce VR sickness. VR sickness can cause strong discomfort, decrease users' presence and enjoyment, especially in games, shorten the duration of the VR experience, and can even pose health risks. Previous research has explored different VR sickness mitigation methods by adding visual effects or elements. Field of View (FOV) reduction, Depth of Field (DOF) blurring, and adding a rest frame into the virtual environment are examples of such methods. Although useful in some cases, they might result in information loss. This research is the first to compare VR sickness, presence, workload to complete a search task, and information loss of these three VR sickness mitigation methods in a racing game with two levels of control. To do this, we conducted a mixed factorial user study (N = 32) with degree of control as the between-subjects factor and the VR sickness mitigation techniques as the within-subjects factor. Participants were required to find targets with three difficulty levels while steering or not steering a car in a virtual environment. Our results show that there are no significant differences in VR sickness, presence and workload among these techniques under two levels of control in our VR racing game. We also found that changing FOV dynamically or using DOF blur effects would result in information loss while adding a target reticule as a rest frame would not.

cs.HC↗

arXiv2021

Developing Real-time Streaming Transformer Transducer for Speech Recognition on Large-scale Dataset

Recently, Transformer based end-to-end models have achieved great success in many areas including speech recognition. However, compared to LSTM models, the heavy computational cost of the Transformer during inference is a key issue to prevent their applications. In this work, we explored the potential of Transformer Transducer (T-T) models for the fist pass decoding with low latency and fast speed on a large-scale dataset. We combine the idea of Transformer-XL and chunk-wise streaming processing to design a streamable Transformer Transducer model. We demonstrate that T-T outperforms the hybrid model, RNN Transducer (RNN-T), and streamable Transformer attention-based encoder-decoder model in the streaming scenario. Furthermore, the runtime cost and latency can be optimized with a relatively small look-ahead.

cs.CL↗

arXiv2021

Learning Audio-Visual Correlations from Variational Cross-Modal Generation

People can easily imagine the potential sound while seeing an event. This natural synchronization between audio and visual signals reveals their intrinsic correlations. To this end, we propose to learn the audio-visual correlations from the perspective of cross-modal generation in a self-supervised manner, the learned correlations can be then readily applied in multiple downstream tasks such as the audio-visual cross-modal localization and retrieval. We introduce a novel Variational AutoEncoder (VAE) framework that consists of Multiple encoders and a Shared decoder (MS-VAE) with an additional Wasserstein distance constraint to tackle the problem. Extensive experiments demonstrate that the optimized latent representation of the proposed MS-VAE can effectively learn the audio-visual correlations and can be readily applied in multiple audio-visual downstream tasks to achieve competitive performance even without any given label information during training.

cs.CV↗

arXiv2021

Learning to Anticipate Egocentric Actions by Imagination

Anticipating actions before they are executed is crucial for a wide range of practical applications, including autonomous driving and robotics. In this paper, we study the egocentric action anticipation task, which predicts future action seconds before it is performed for egocentric videos. Previous approaches focus on summarizing the observed content and directly predicting future action based on past observations. We believe it would benefit the action anticipation if we could mine some cues to compensate for the missing information of the unobserved frames. We then propose to decompose the action anticipation into a series of future feature predictions. We imagine how the visual feature changes in the near future and then predicts future action labels based on these imagined representations. Differently, our ImagineRNN is optimized in a contrastive learning way instead of feature regression. We utilize a proxy task to train the ImagineRNN, i.e., selecting the correct future states from distractors. We further improve ImagineRNN by residual anticipation, i.e., changing its target to predicting the feature difference of adjacent frames instead of the frame content. This promotes the network to focus on our target, i.e., the future action, as the difference between adjacent frame features is more important for forecasting the future. Extensive experiments on two large-scale egocentric action datasets validate the effectiveness of our method. Our method significantly outperforms previous methods on both the seen test set and the unseen test set of the EPIC Kitchens Action Anticipation Challenge.

cs.CV↗

arXiv2020

Average Doppler Shift of Gamma-ray Spectra of Positron Annihilation Process in Molecules

This paper studies the gamma-ray spectra of positron annihilation processes in a series of molecules. The results show that the average valence electron energy of the molecules has a linear correlation with the full width at half maximum (FWHM) of the gamma-ray spectra. In addition, we defined a new physical quantity Average Doppler Shift (ADS), which can be used as the eigenvalue to describe the characteristics of the gamma-ray spectra. Since ADS contains all the information about the gamma-ray spectra, it can more accurately represent the characteristics of the gamma-ray spectra. For a series of molecules, this paper compares the ADS and FWHM of their gamma-ray spectra and the average valence electron energy. The results show that ADS has a linear correlation with the average valence electron energy and the FWHM. Further, this proves that the annihilation mainly occurs on valence electrons, and it also illustrates that the ADS has certain applicability. It is expected that this will provide us with a deeper understanding of the positron annihilation process.

physics.atm-clus↗