no code implementations • EMNLP 2021 • Tao Ji, Yong Jiang, Tao Wang, Zhongqiang Huang, Fei Huang, Yuanbin Wu, Xiaoling Wang
Adapting word order from one language to another is a key problem in cross-lingual structured prediction.
no code implementations • EMNLP 2021 • Tao Ji, Yong Jiang, Tao Wang, Zhongqiang Huang, Fei Huang, Yuanbin Wu, Xiaoling Wang
Transition systems usually contain various dynamic structures (e. g., stacks, buffers).
no code implementations • 27 Mar 2022 • Neng Wang, Yang Bai, Kun Yu, Yong Jiang, Shu-Tao Xia, Yan Wang
Face forgery has attracted increasing attention in recent applications of computer vision.
1 code implementation • 1 Mar 2022 • Xinyu Wang, Yongliang Shen, Jiong Cai, Tao Wang, Xiaobin Wang, Pengjun Xie, Fei Huang, Weiming Lu, Yueting Zhuang, Kewei Tu, Wei Lu, Yong Jiang
Our system wins 10 out of 13 tracks in the MultiCoNER shared task.
Multilingual Named Entity Recognition
named-entity-recognition
+1
1 code implementation • 8 Feb 2022 • Yue He, Zimu Wang, Peng Cui, Hao Zou, Yafeng Zhang, Qiang Cui, Yong Jiang
In spite of the tremendous development of recommender system owing to the progressive capability of machine learning recently, the current recommender system is still vulnerable to the distribution shift of users and items in realistic scenarios, leading to the sharp decline of performance in testing environments.
1 code implementation • ICLR 2022 • Yiming Li, Haoxiang Zhong, Xingjun Ma, Yong Jiang, Shu-Tao Xia
Visual object tracking (VOT) has been widely adopted in mission-critical applications, such as autonomous driving and intelligent surveillance systems.
1 code implementation • 13 Dec 2021 • Xinyu Wang, Min Gui, Yong Jiang, Zixia Jia, Nguyen Bach, Tao Wang, Zhongqiang Huang, Fei Huang, Kewei Tu
As text representations take the most important role in MNER, in this paper, we propose {\bf I}mage-{\bf t}ext {\bf A}lignments (ITA) to align image features into the textual space, so that the attention mechanism in transformer-based pretrained textual embeddings can be better utilized.
1 code implementation • ICML Workshop AML 2021 • Yiming Li, Linghui Zhu, Xiaojun Jia, Yong Jiang, Shu-Tao Xia, Xiaochun Cao
In this paper, we explore the defense from another angle by verifying whether a suspicious model contains the knowledge of defender-specified \emph{external features}.
no code implementations • NeurIPS 2021 • Yang Bai, Xin Yan, Yong Jiang, Shu-Tao Xia, Yisen Wang
Adversarial robustness has received increasing attention along with the study of adversarial examples.
1 code implementation • 25 Nov 2021 • Yang Bai, Xin Yan, Yong Jiang, Shu-Tao Xia, Yisen Wang
Adversarial robustness has received increasing attention along with the study of adversarial examples.
1 code implementation • 13 Oct 2021 • Yu Zhang, Qingrong Xia, Shilin Zhou, Yong Jiang, Zhenghua Li, Guohong Fu, Min Zhang
Semantic role labeling is a fundamental yet challenging task in the NLP community.
Ranked #1 on
Semantic Role Labeling
on OntoNotes
no code implementations • 8 Oct 2021 • Ke Zhang, Sihong Chen, Qi Ju, Yong Jiang, Yucong Li, Xin He
The graph network that is established with patches as the nodes can maximize the mutual learning of similar objects.
no code implementations • 29 Sep 2021 • Naiqi Li, Wenjie Li, Yong Jiang, Shu-Tao Xia
In this paper we propose the deep Dirichlet process mixture (DDPM) model, which is an unsupervised method that simultaneously performs clustering and feature learning.
1 code implementation • EMNLP 2021 • Xinyin Ma, Yong Jiang, Nguyen Bach, Tao Wang, Zhongqiang Huang, Fei Huang, Weiming Lu
Entity retrieval, which aims at disambiguating mentions to canonical entities from massive KBs, is essential for many tasks in natural language processing.
Ranked #1 on
Entity Retrieval
on ZESHEL
no code implementations • 9 Aug 2021 • Huimin Zhou, Qing Li, Yong Jiang, Rongwei Yang, Zhuyun Qi
In the current deep learning based recommendation system, the embedding method is generally employed to complete the conversion from the high-dimensional sparse feature vector to the low-dimensional dense feature vector.
no code implementations • ACL 2021 • Zechuan Hu, Yong Jiang, Nguyen Bach, Tao Wang, Zhongqiang Huang, Fei Huang, Kewei Tu
In this paper, we propose a novel unified framework for zero-shot sequence labeling with minimum risk training and design a new decomposable risk function that models the relations between the predicted labels from the source models and the true labels.
no code implementations • ACL 2021 • Zechuan Hu, Yong Jiang, Nguyen Bach, Tao Wang, Zhongqiang Huang, Fei Huang, Kewei Tu
In structured prediction problems, cross-lingual transfer learning is an efficient way to train quality models for low-resource languages, and further improvement can be obtained by learning from multiple source languages.
no code implementations • 7 Jul 2021 • Peidong Liu, Zibin He, Xiyu Yan, Yong Jiang, Shutao Xia, Feng Zheng, Maowei Hu
In this work, we propose an effective weakly-supervised video semantic segmentation pipeline with click annotations, called WeClick, for saving laborious annotating effort by segmenting an instance of the semantic class with only a single click.
no code implementations • ACL (IWPT) 2021 • Xinyu Wang, Zixia Jia, Yong Jiang, Kewei Tu
This paper describes the system used in submission from SHANGHAITECH team to the IWPT 2021 Shared Task.
1 code implementation • ACL 2021 • Siyang Liu, Chujie Zheng, Orianna Demasi, Sahand Sabour, Yu Li, Zhou Yu, Yong Jiang, Minlie Huang
Emotional support is a crucial ability for many conversation scenarios, including social interactions, mental health support, and customer service chats.
1 code implementation • ACL 2021 • Yida Wang, Yinhe Zheng, Yong Jiang, Minlie Huang
Neural dialogue generation models trained with the one-hot target distribution suffer from the over-confidence issue, which leads to poor generation diversity as widely reported in the literature.
1 code implementation • ACL 2021 • Xinyu Wang, Yong Jiang, Nguyen Bach, Tao Wang, Zhongqiang Huang, Fei Huang, Kewei Tu
We find empirically that the contextual representations computed on the retrieval-based input view, constructed through the concatenation of a sentence and its external contexts, can achieve significantly improved performance compared to the original input view based only on the sentence.
Ranked #1 on
Named Entity Recognition
on WNUT 2017
no code implementations • 6 Apr 2021 • Yiming Li, Tongqing Zhai, Yong Jiang, Zhifeng Li, Shu-Tao Xia
We demonstrate that this attack paradigm is vulnerable when the trigger in testing images is not consistent with the one used for training.
no code implementations • EACL 2021 • Kewei Tu, Yong Jiang, Wenjuan Han, Yanpeng Zhao
Unsupervised parsing learns a syntactic parser from training sentences without parse tree annotations.
1 code implementation • 28 Mar 2021 • Rui Zhang, Bayu Distiawan Trisedy, Miao Li, Yong Jiang, Jianzhong Qi
In the last few years, the interest in knowledge bases has grown exponentially in both the research community and the industry due to their essential role in AI applications.
1 code implementation • ICLR 2021 • Yang Bai, Yuyuan Zeng, Yong Jiang, Shu-Tao Xia, Xingjun Ma, Yisen Wang
The study of adversarial examples and their activation has attracted significant attention for secure and robust learning with deep neural networks (DNNs).
no code implementations • 6 Mar 2021 • Yiming Li, YanJie Li, Yalei Lv, Yong Jiang, Shu-Tao Xia
Deep neural networks (DNNs) are vulnerable to the \emph{backdoor attack}, which intends to embed hidden backdoors in DNNs by poisoning training data.
1 code implementation • ICLR 2021 • Peidong Liu, Gengwei Zhang, Bochao Wang, Hang Xu, Xiaodan Liang, Yong Jiang, Zhenguo Li
For object detection, the well-established classification and regression loss functions have been carefully designed by considering diverse learning challenges.
1 code implementation • 3 Dec 2020 • Han Qiu, Yi Zeng, Tianwei Zhang, Yong Jiang, Meikang Qiu
With more and more advanced adversarial attack methods have been developed, a quantity of corresponding defense solutions were designed to enhance the robustness of DNN models.
1 code implementation • NeurIPS 2020 • Naiqi Li, Wenjie Li, Jifeng Sun, Yinghua Gao, Yong Jiang, Shu-Tao Xia
In this paper we propose Stochastic Deep Gaussian Processes over Graphs (DGPG), which are deep structure models that learn the mappings between input and output signals in graph domains.
no code implementations • NeurIPS 2020 • Chaobing Song, Zhengyuan Zhou, Yichao Zhou, Yong Jiang, Yi Ma
The optimization problems associated with training generative adversarial neural networks can be largely reduced to certain {\em non-monotone} variational inequality problems (VIPs), whereas existing convergence results are mostly based on monotone or strongly monotone assumptions.
no code implementations • Findings of the Association for Computational Linguistics 2020 • Zechuan Hu, Yong Jiang, Nguyen Bach, Tao Wang, Zhongqiang Huang, Fei Huang, Kewei Tu
The neural linear-chain CRF model is one of the most widely-used approach to sequence labeling.
no code implementations • 10 Nov 2020 • Yang Zhou, Yong Jiang, Zechuan Hu, Kewei Tu
One limitation of linear chain CRFs is their inability to model long-range dependencies between labels.
no code implementations • 2 Nov 2020 • Paul Voigtlaender, Lishu Luo, Chun Yuan, Yong Jiang, Bastian Leibe
We use a deep convolutional network to automatically create pseudo-labels on a pixel level from much cheaper bounding box annotations and investigate how far such pseudo-labels can carry us for training state-of-the-art VOS approaches.
1 code implementation • COLING 2020 • Songlin Yang, Yong Jiang, Wenjuan Han, Kewei Tu
Inspired by second-order supervised dependency parsing, we proposed a second-order extension of unsupervised neural dependency models that incorporate grandparent-child or sibling information.
Ranked #1 on
Dependency Grammar Induction
on WSJ10
1 code implementation • 22 Oct 2020 • Tongqing Zhai, Yiming Li, Ziqi Zhang, Baoyuan Wu, Yong Jiang, Shu-Tao Xia
We also demonstrate that existing backdoor attacks cannot be directly adopted in attacking speaker verification.
1 code implementation • 12 Oct 2020 • Yiming Li, Ziqi Zhang, Jiawang Bai, Baoyuan Wu, Yong Jiang, Shu-Tao Xia
Based on the proposed backdoor-based watermarking, we use a hypothesis test guided method for dataset verification based on the posterior probability generated by the suspicious third-party model of the benign samples and their correspondingly watermarked samples ($i. e.$, images with trigger) on the target class.
1 code implementation • ACL 2021 • Xinyu Wang, Yong Jiang, Nguyen Bach, Tao Wang, Zhongqiang Huang, Fei Huang, Kewei Tu
Pretrained contextualized embeddings are powerful word representations for structured prediction tasks.
Ranked #1 on
Part-Of-Speech Tagging
on ARK
1 code implementation • ACL 2021 • Xinyu Wang, Yong Jiang, Zhaohui Yan, Zixia Jia, Nguyen Bach, Tao Wang, Zhongqiang Huang, Fei Huang, Kewei Tu
The objective function of knowledge distillation is typically the cross-entropy between the teacher and the student's output distributions.
1 code implementation • EMNLP 2020 • Wenjuan Han, Liwen Zhang, Yong Jiang, Kewei Tu
To address these problems, we propose a novel and unified framework that learns to attack a structured prediction model using a sequence-to-sequence model with feedbacks from multiple reference models of the same structured prediction task.
no code implementations • COLING 2020 • Wenjuan Han, Yong Jiang, Hwee Tou Ng, Kewei Tu
Syntactic dependency parsing is an important task in natural language processing.
1 code implementation • ECCV 2020 • Yang Bai, Yuyuan Zeng, Yong Jiang, Yisen Wang, Shu-Tao Xia, Weiwei Guo
Deep neural networks (DNNs) have demonstrated excellent performance on various tasks, however they are under the risk of adversarial examples that can be easily generated when the target model is accessible to an attacker (white-box setting).
1 code implementation • EMNLP 2020 • Xinyu Wang, Yong Jiang, Nguyen Bach, Tao Wang, Zhongqiang Huang, Fei Huang, Kewei Tu
The linear-chain Conditional Random Field (CRF) model is one of the most widely-used neural sequence labeling approaches.
Ranked #3 on
Chunking
on CoNLL 2003 (German)
no code implementations • Findings of the Association for Computational Linguistics 2020 • Xinyu Wang, Yong Jiang, Nguyen Bach, Tao Wang, Zhongqiang Huang, Fei Huang, Kewei Tu
Recent work proposes a family of contextual embeddings that significantly improves the accuracy of sequence labelers over non-contextual embeddings.
Ranked #2 on
Chunking
on CoNLL 2003 (German)
no code implementations • 21 Aug 2020 • Yiming Li, Jiawang Bai, Jiawei Li, Xue Yang, Yong Jiang, Shu-Tao Xia
Interpretability and effectiveness are two essential and indispensable requirements for adopting machine learning methods in reality.
no code implementations • 14 Aug 2020 • Jie Fang, Jian-Wu Lin, Shu-Tao Xia, Yong Jiang, Zhikang Xia, Xiang Liu
This paper proposes Neural Network-based Automatic Factor Construction (NNAFC), a tailored neural network framework that can automatically construct diversified financial factors based on financial domain knowledge and a variety of neural network structures.
3 code implementations • 10 Aug 2020 • Yida Wang, Pei Ke, Yinhe Zheng, Kaili Huang, Yong Jiang, Xiaoyan Zhu, Minlie Huang
The cleaned dataset and the pre-training models will facilitate the research of short-text conversation modeling.
1 code implementation • 17 Jul 2020 • Yiming Li, Yong Jiang, Zhifeng Li, Shu-Tao Xia
Backdoor attack intends to embed hidden backdoor into deep neural networks (DNNs), so that the attacked models perform well on benign samples, whereas their predictions will be maliciously changed if the hidden backdoor is activated by attacker-specified triggers.
no code implementations • ACL 2020 • Jun Li, Yifan Cao, Jiong Cai, Yong Jiang, Kewei Tu
Unsupervised constituency parsing aims to learn a constituency parser from a training corpus without parse tree annotations.
no code implementations • NeurIPS 2020 • Chaobing Song, Yong Jiang, Yi Ma
Meanwhile, VRADA matches the lower bound of the general convex setting up to a $\log\log n$ factor and matches the lower bounds in both regimes $n\le \Theta(\kappa)$ and $n\gg \kappa$ of the strongly convex setting, where $\kappa$ denotes the condition number.
1 code implementation • WS 2020 • Xinyu Wang, Yong Jiang, Kewei Tu
This paper presents the system used in our submission to the \textit{IWPT 2020 Shared Task}.
no code implementations • 9 Apr 2020 • Yiming Li, Tongqing Zhai, Baoyuan Wu, Yong Jiang, Zhifeng Li, Shu-Tao Xia
Backdoor attack intends to inject hidden backdoor into the deep neural networks (DNNs), such that the prediction of the infected model will be maliciously changed if the hidden backdoor is activated by the attacker-defined trigger, while it performs well on benign samples.
1 code implementation • ACL 2020 • Xinyu Wang, Yong Jiang, Nguyen Bach, Tao Wang, Fei Huang, Kewei Tu
Multilingual sequence labeling is a task of predicting label sequences using a single unified model for multiple languages.
1 code implementation • 16 Mar 2020 • Yiming Li, Baoyuan Wu, Yan Feng, Yanbo Fan, Yong Jiang, Zhifeng Li, Shu-Tao Xia
In this work, we propose a novel defense method, the robust training (RT), by jointly minimizing two separated risks ($R_{stand}$ and $R_{rob}$), which is with respect to the benign example and its neighborhoods respectively.
no code implementations • 26 Dec 2019 • Jie Fang, Shu-Tao Xia, Jian-Wu Lin, Zhikang Xia, Xiang Liu, Yong Jiang
This paper proposes Alpha Discovery Neural Network (ADNN), a tailored neural network structure which can automatically construct diversified financial technical indicators based on prior knowledge.
no code implementations • 8 Dec 2019 • Jie Fang, Shu-Tao Xia, Jian-Wu Lin, Yong Jiang
According to neural network universal approximation theorem, pre-training can conduct a more effective and explainable evolution process.
1 code implementation • 5 Nov 2019 • Yiming Li, Peidong Liu, Yong Jiang, Shu-Tao Xia
To a large extent, the privacy of visual classification data is mainly in the mapping between the image and its corresponding label, since this relation provides a great amount of information and can be used in other scenarios.
no code implementations • 5 Nov 2019 • Peidong Liu, Xiyu Yan, Yong Jiang, Shu-Tao Xia
The deep learning-based visual tracking algorithms such as MDNet achieve high performance leveraging to the feature extraction ability of a deep neural network.
no code implementations • IJCNLP 2019 • Yong Jiang, Wenjuan Han, Kewei Tu
Grammar induction aims to discover syntactic structures from unannotated sentences.
no code implementations • IJCNLP 2019 • Wenjuan Han, Ge Wang, Yong Jiang, Kewei Tu
The key to multilingual grammar induction is to couple grammar parameters of different languages together by exploiting the similarity between languages.
no code implementations • 27 Oct 2019 • Jia Xu, Yiming Li, Yong Jiang, Shu-Tao Xia
In this paper, we define the local flatness of the loss surface as the maximum value of the chosen norm of the gradient regarding to the input within a neighborhood centered on the benign sample, and discuss the relationship between the local flatness and adversarial vulnerability.
1 code implementation • AAAI 2019 • Yunzhe Yuan, Yong Jiang, Kewei Tu
Traditionally, a transitionbased dependency parser processes an input sentence and predicts a sequence of parsing actions in a left-to-right manner.
1 code implementation • 17 Jul 2019 • Yiming Li, Yang Zhang, Qingtao Tang, Weipeng Huang, Yong Jiang, Shu-Tao Xia
$k$-means algorithm is one of the most classical clustering methods, which has been widely and successfully used in signal processing.
no code implementations • ACL 2019 • Wenjuan Han, Yong Jiang, Kewei Tu
In this paper, we propose a novel probabilistic model called discriminative neural dependency model with valence (D-NDMV) that generates a sentence and its parse from a continuous latent representation, which encodes global contextual information of the generated sentence.
Ranked #2 on
Dependency Grammar Induction
on WSJ10
Constituency Grammar Induction
Dependency Grammar Induction
+1
no code implementations • WS 2019 • Shaobo Cui, Rongzhong Lian, Di Jiang, Yuanfeng Song, Siqi Bao, Yong Jiang
DAL is the first work to innovatively utilizes the duality between query generation and response generation to avoid safe responses and increase the diversity of the generated responses.
no code implementations • 3 Jun 2019 • Chaobing Song, Yong Jiang, Yi Ma
In this general convex setting, we propose a concise unified acceleration framework (UAF), which reconciles the two different high-order acceleration approaches, one by Nesterov and Baes [29, 3, 33] and one by Monteiro and Svaiter [25].
no code implementations • 14 Mar 2019 • Jiawang Bai, Yiming Li, Jiawei Li, Yong Jiang, Shu-Tao Xia
How to obtain a model with good interpretability and performance has always been an important research topic.
no code implementations • 10 Mar 2019 • Yiming Li, Jiawang Bai, Jiawei Li, Xue Yang, Yong Jiang, Chun Li, Shu-Tao Xia
Despite the impressive performance of random forests (RF), its theoretical properties have not been thoroughly understood.
no code implementations • 25 Sep 2018 • Chaobing Song, Ji Liu, Han Liu, Yong Jiang, Tong Zhang
Regularized online learning is widely used in machine learning applications.
no code implementations • 21 Apr 2018 • Zhaopeng Tu, Yong Jiang, Xiaojiang Liu, Lei Shu, Shuming Shi
We study the problem of stock related question answering (StockQA): automatically generating answers to stock related questions, just like professional stock analysts providing action recommendations to stocks upon user's requests.
1 code implementation • EMNLP 2017 • Xiao Zhang, Yong Jiang, Hao Peng, Kewei Tu, Dan Goldwasser
In this paper we propose an end-to-end neural CRF autoencoder (NCRF-AE) model for semi-supervised learning of sequential structured prediction problems.
no code implementations • 16 Aug 2017 • Jun Mei, Yong Jiang, Kewei Tu
For the theoretical part, we reduce general MAP inference to its special case without evidence and hidden variables; we also show that it is NP-hard to approximate the MAP problem to $2^{n^\epsilon}$ for fixed $0 \leq \epsilon < 1$, where $n$ is the input size.
1 code implementation • EMNLP 2017 • Jiong Cai, Yong Jiang, Kewei Tu
The encoder part of our model is discriminative and globally normalized which allows us to use rich features as well as universal linguistic priors.
Dependency Grammar Induction
Unsupervised Dependency Parsing
no code implementations • EMNLP 2017 • Wenjuan Han, Yong Jiang, Kewei Tu
We study the impact of big models (in terms of the degree of lexicalization) and big data (in terms of the training corpus size) on dependency grammar induction.
no code implementations • EMNLP 2017 • Yong Jiang, Wenjuan Han, Kewei Tu
Unsupervised dependency parsing aims to learn a dependency parser from unannotated sentences.
Dependency Grammar Induction
Unsupervised Dependency Parsing
no code implementations • 8 Sep 2016 • Shanbo Chu, Yong Jiang, Kewei Tu
Probabilistic modeling is one of the foundations of modern machine learning and artificial intelligence.