no code implementations • 17 Dec 2013 • Yalong Bai, Kuiyuan Yang, Wei Yu, Wei-Ying Ma, Tiejun Zhao
Image retrieval refers to finding relevant images from an image database for a query, which is considered difficult for the gap between low-level representation of images and high-level representation of queries.
no code implementations • 1 Dec 2015 • Yiming Cui, Conghui Zhu, Xiaoning Zhu, Tiejun Zhao
Pivot language is employed as a way to solve the data sparseness problem in machine translation, especially when the data for a particular language pair does not exist.
no code implementations • LREC 2016 • Huaxing Shi, Tiejun Zhao, Keh-Yih Su
This Treebank is a part of a semantic corpus building project, which aims to build a semantic bilingual corpus annotated with syntactic, semantic cases and word senses.
no code implementations • COLING 2016 • Hailong Cao, Tiejun Zhao, Shu Zhang, Yao Meng
We introduce a distribution based model to learn bilingual word embeddings from monolingual data.
1 code implementation • COLING 2016 • Junwei Bao, Nan Duan, Zhao Yan, Ming Zhou, Tiejun Zhao
WebQuestions and SimpleQuestions are two benchmark data-sets commonly used in recent knowledge-based question answering (KBQA) work.
no code implementations • 28 Aug 2017 • Yalong Bai, Kuiyuan Yang, Tao Mei, Wei-Ying Ma, Tiejun Zhao
Large scale image dataset and deep convolutional neural network (DCNN) are two primary driving forces for the rapid progress made in generic object recognition tasks in recent years.
no code implementations • EMNLP 2017 • Kehai Chen, Rui Wang, Masao Utiyama, Lemao Liu, Akihiro Tamura, Eiichiro Sumita, Tiejun Zhao
Source dependency information has been successfully introduced into statistical machine translation.
no code implementations • IJCNLP 2017 • Kehai Chen, Rui Wang, Masao Utiyama, Eiichiro Sumita, Tiejun Zhao
In Neural Machine Translation (NMT), each word is represented as a low-dimension, real-value vector for encoding its syntax and semantic information.
no code implementations • 12 Nov 2017 • Kehai Chen, Rui Wang, Masao Utiyama, Eiichiro Sumita, Tiejun Zhao
In this paper, we extend local attention with syntax-distance constraint, to focus on syntactically related source words with the predicted target word, thus learning a more effective context vector for word prediction.
no code implementations • 29 May 2018 • Junwei Bao, Duyu Tang, Nan Duan, Zhao Yan, Yuanhua Lv, Ming Zhou, Tiejun Zhao
The model maps a row from a table to a continuous vector and then generates a natural language sentence by leveraging the semantics of a table.
no code implementations • ACL 2018 • Chunpeng Ma, Akihiro Tamura, Masao Utiyama, Tiejun Zhao, Eiichiro Sumita
Tree-based neural machine translation (NMT) approaches, although achieved impressive performance, suffer from a major drawback: they only use the 1-best parse tree to direct the translation, which potentially introduces translation mistakes due to parsing errors.
1 code implementation • ACL 2018 • Qingyu Zhou, Nan Yang, Furu Wei, Shaohan Huang, Ming Zhou, Tiejun Zhao
In this paper, we present a novel end-to-end neural network framework for extractive document summarization by jointly learning to score and select sentences.
Ranked #9 on Extractive Text Summarization on CNN / Daily Mail
no code implementations • ECCV 2018 • Yalong Bai, Jianlong Fu, Tiejun Zhao, Tao Mei
First, we model one of the pairwise interaction (e. g., image and question) by bilinear features, which is further encoded with the third dimension (e. g., answer) to be a triplet by bilinear tensor product.
2 code implementations • ACL 2019 • Guanhua Zhang, Bing Bai, Jian Liang, Kun Bai, Shiyu Chang, Mo Yu, Conghui Zhu, Tiejun Zhao
Natural Language Sentence Matching (NLSM) has gained substantial attention from both academics and the industry, and rich public datasets contribute a lot to this process.
no code implementations • NAACL 2019 • Chunpeng Ma, Akihiro Tamura, Masao Utiyama, Eiichiro Sumita, Tiejun Zhao
The explicit use of syntactic information has been proved useful for neural machine translation (NMT).
no code implementations • NAACL 2019 • Guanlin Li, Lemao Liu, Xintong Li, Conghui Zhu, Tiejun Zhao, Shuming Shi
Multilayer architectures are currently the gold standard for large-scale neural machine translation.
no code implementations • SEMEVAL 2019 • Yaojie Zhang, Bing Xu, Tiejun Zhao
Our macro-averaged F1-score in sub-task A is 0. 768, ranking 28/103.
no code implementations • ACL 2019 • Mingming Yang, Rui Wang, Kehai Chen, Masao Utiyama, Eiichiro Sumita, Min Zhang, Tiejun Zhao
The training objective of neural machine translation (NMT) is to minimize the loss between the words in the translated sentences and those in the references.
no code implementations • ACL 2019 • Haipeng Sun, Rui Wang, Kehai Chen, Masao Utiyama, Eiichiro Sumita, Tiejun Zhao
In previous methods, UBWE is first trained using non-parallel monolingual corpora and then this pre-trained UBWE is used to initialize the word embedding in the encoder and decoder of UNMT.
no code implementations • 26 Aug 2019 • Haipeng Sun, Rui Wang, Kehai Chen, Masao Utiyama, Eiichiro Sumita, Tiejun Zhao, Chenhui Chu
However, it has not been well-studied for unsupervised neural machine translation (UNMT), although UNMT has recently achieved remarkable results in several domain-specific language pairs.
no code implementations • 3 Sep 2019 • Xuefeng Bai, Yue Zhang, Hailong Cao, Tiejun Zhao
Unsupervised bilingual lexicon induction naturally exhibits duality, which results from symmetry in back-translation.
no code implementations • 10 Sep 2019 • Guanhua Zhang, Bing Bai, Junqi Zhang, Kun Bai, Conghui Zhu, Tiejun Zhao
This irregularity makes the evaluation results over-estimated and affects models' generalization ability.
no code implementations • IJCNLP 2019 • Guanlin Li, Lemao Liu, Guoping Huang, Conghui Zhu, Tiejun Zhao
Many Data Augmentation (DA) methods have been proposed for neural machine translation.
no code implementations • 7 Feb 2020 • Chaoqun Duan, Lei Cui, Shuming Ma, Furu Wei, Conghui Zhu, Tiejun Zhao
In this work, we aim to improve the relevance between live comments and videos by modeling the cross-modal interactions among different modalities.
no code implementations • COLING 2020 • Haipeng Sun, Rui Wang, Kehai Chen, Xugang Lu, Masao Utiyama, Eiichiro Sumita, Tiejun Zhao
Unsupervised neural machine translation (UNMT) has recently attracted great interest in the machine translation community.
no code implementations • 28 Feb 2020 • Chaoqun Duan, Kehai Chen, Rui Wang, Masao Utiyama, Eiichiro Sumita, Conghui Zhu, Tiejun Zhao
Existing neural machine translation (NMT) systems utilize sequence-to-sequence neural networks to generate target translation word by word, and then make the generated word at each time-step and the counterpart in the references as consistent as possible.
2 code implementations • CVPR 2020 • Mohan Zhou, Yalong Bai, Wei zhang, Tiejun Zhao, Tao Mei
Specifically, we first propose an object-extent learning module for localizing the object according to the visual patterns shared among the instances in the same category.
Ranked #17 on Fine-Grained Image Classification on CUB-200-2011
no code implementations • 5 Apr 2020 • Guanlin Li, Lemao Liu, Conghui Zhu, Tiejun Zhao, Shuming Shi
Generalization to unseen instances is our eternal pursuit for all data-driven models.
no code implementations • 5 Apr 2020 • Conghui Zhu, Guanlin Li, Lemao Liu, Tiejun Zhao, Shuming Shi
Despite the great success of NMT, there still remains a severe challenge: it is hard to interpret the internal dynamics during its training process.
no code implementations • NAACL 2021 • Haipeng Sun, Rui Wang, Kehai Chen, Masao Utiyama, Eiichiro Sumita, Tiejun Zhao
Unsupervised neural machine translation (UNMT) that relies solely on massive monolingual corpora has achieved remarkable results in several translation tasks.
no code implementations • ACL 2020 • Haipeng Sun, Rui Wang, Kehai Chen, Masao Utiyama, Eiichiro Sumita, Tiejun Zhao
Unsupervised neural machine translation (UNMT) has recently achieved remarkable results for several language pairs.
1 code implementation • ACL 2020 • Guanhua Zhang, Bing Bai, Junqi Zhang, Kun Bai, Conghui Zhu, Tiejun Zhao
In this paper, we formalize the unintended biases in text classification datasets as a kind of selection bias from the non-discrimination distribution to the discrimination distribution.
no code implementations • WS 2020 • Xuancai Li, Chen Kehai, Tiejun Zhao, Muyun Yang
End-to-End speech translation usually leverages audio-to-text parallel data to train an available speech translation model which has shown impressive results on various speech translation tasks.
no code implementations • 15 Oct 2020 • Guanhua Zhang, Bing Bai, Jian Liang, Kun Bai, Conghui Zhu, Tiejun Zhao
Recent studies show that crowd-sourced Natural Language Inference (NLI) datasets may suffer from significant biases like annotation artifacts.
no code implementations • COLING 2020 • Yingyao Wang, Junwei Bao, Guangyi Liu, Youzheng Wu, Xiaodong He, BoWen Zhou, Tiejun Zhao
This paper aims to enhance the few-shot relation classification especially for sentences that jointly describe multiple relations.
1 code implementation • EMNLP 2020 • Changzhen Ji, Xin Zhou, Yating Zhang, Xiaozhong Liu, Changlong Sun, Conghui Zhu, Tiejun Zhao
In the past few years, audiences from different fields witness the achievements of sequence-to-sequence models (e. g., LSTM+attention, Pointer Generator Networks, and Transformer) to enhance dialogue content generation.
no code implementations • 22 Oct 2020 • Changzhen Ji, Xin Zhou, Conghui Zhu, Tiejun Zhao
The multi-role judicial debate composed of the plaintiff, defendant, and judge is an important part of the judicial trial.
no code implementations • SEMEVAL 2020 • Zhen Li, Yaojie Zhang, Bing Xu, Tiejun Zhao
Internet memes emotion recognition is focused by many researchers.
no code implementations • COLING 2020 • Zhenyu Zhao, Shuangzhi Wu, Muyun Yang, Kehai Chen, Tiejun Zhao
Neural models have achieved great success on the task of machine reading comprehension (MRC), which are typically trained on hard labels.
1 code implementation • 21 Dec 2020 • Wang Xu, Kehai Chen, Tiejun Zhao
In document-level relation extraction (DocRE), graph structure is generally used to encode relation information in the input document to classify the relation category between each entity pair, and has greatly advanced the DocRE task over the past several years.
Ranked #35 on Relation Extraction on DocRED
no code implementations • 1 Jan 2021 • Guanlin Li, Lemao Liu, Taro Watanabe, Conghui Zhu, Tiejun Zhao
Unsupervised Neural Machine Translation or UNMT has received great attention in recent years.
no code implementations • 16 Mar 2021 • Pengbo Liu, Hailong Cao, Tiejun Zhao
Multi-modal machine translation (MMT) improves translation quality by introducing visual information.
Ranked #5 on Multimodal Machine Translation on Multi30K
no code implementations • 26 May 2021 • Hailong Cao, Tiejun Zhao
Embeddings of two languages are made to match with each other by rotating and scaling.
2 code implementations • Findings (ACL) 2021 • Wang Xu, Kehai Chen, Tiejun Zhao
Document-level relation extraction (DocRE) models generally use graph networks to implicitly model the reasoning skill (i. e., pattern recognition, logical reasoning, coreference reasoning, etc.)
Ranked #24 on Relation Extraction on DocRED
no code implementations • SEMEVAL 2021 • Chenyi Wang, Tianshu Liu, Tiejun Zhao
This paper introduces our system at SemEval-2021 Task 5: Toxic Spans Detection.
1 code implementation • ACL 2021 • Tingting Ma, Jin-Ge Yao, Chin-Yew Lin, Tiejun Zhao
The general format of natural language inference (NLI) makes it tempting to be used for zero-shot text classification by casting any target label into a sentence of hypothesis and verifying whether or not it could be entailed by the input, aiming at generic classification applicable on any specified label space.
1 code implementation • 18 Aug 2021 • Yongwei Zhou, Junwei Bao, Haipeng Sun, Jiahui Liang, Youzheng Wu, Xiaodong He, BoWen Zhou, Tiejun Zhao
Reasoning machine reading comprehension (R-MRC) aims to answer complex questions that require discrete reasoning based on text.
1 code implementation • 20 Aug 2021 • Changzhen Ji, Yating Zhang, Xiaozhong Liu, Adam Jatowt, Changlong Sun, Conghui Zhu, Tiejun Zhao
Nevertheless, few works utilized the knowledge extracted from similar conversations for utterance generation.
no code implementations • 27 Dec 2021 • Mohan Zhou, Yalong Bai, Wei zhang, Ting Yao, Tiejun Zhao, Tao Mei
Automatically synthesizing listening behavior that actively responds to a talking head, is critical to applications such as digital human, virtual agents and social robots.
1 code implementation • Findings (ACL) 2022 • Tingting Ma, Huiqiang Jiang, Qianhui Wu, Tiejun Zhao, Chin-Yew Lin
Few-shot named entity recognition (NER) systems aim at recognizing novel-class named entities based on only a few labeled examples.
Ranked #5 on Few-shot NER on Few-NERD (INTRA)
1 code implementation • Findings (NAACL) 2022 • Zhen Li, Bing Xu, Conghui Zhu, Tiejun Zhao
Compared with unimodal data, multimodal data can provide more features to help the model analyze the sentiment of data.
1 code implementation • NAACL 2022 • Wang Xu, Kehai Chen, Lili Mou, Tiejun Zhao
Document-level relation extraction (DocRE) aims to determine the relation between two entities from a document of multiple sentences.
Ranked #5 on Dialog Relation Extraction on DialogRE (F1c (v1) metric)
Dialog Relation Extraction Document-level Relation Extraction +2
no code implementations • 29 Apr 2022 • Yongwei Zhou, Junwei Bao, Chaoqun Duan, Haipeng Sun, Jiahui Liang, Yifan Wang, Jing Zhao, Youzheng Wu, Xiaodong He, Tiejun Zhao
To inherit the advantages of these two types of methods, we propose OPERA, an operation-pivoted discrete reasoning framework, where lightweight symbolic operations (compared with logical forms) as neural modules are utilized to facilitate the reasoning ability and interpretability.
1 code implementation • 15 Oct 2022 • Yongwei Zhou, Junwei Bao, Chaoqun Duan, Youzheng Wu, Xiaodong He, Tiejun Zhao
Question answering requiring discrete reasoning, e. g., arithmetic computing, comparison, and counting, over knowledge is a challenging task.
1 code implementation • 19 Oct 2022 • Yingyao Wang, Junwei Bao, Chaoqun Duan, Youzheng Wu, Xiaodong He, Tiejun Zhao
To preserve the advantage and eliminate the disadvantage of different granularity evidence, we propose MuGER$^2$, a Multi-Granularity Evidence Retrieval and Reasoning approach.
no code implementations • 20 Oct 2022 • Wanjun Zhong, Tingting Ma, Jiahai Wang, Jian Yin, Tiejun Zhao, Chin-Yew Lin, Nan Duan
This paper presents ReasonFormer, a unified reasoning framework for mirroring the modular and compositional reasoning process of humans in complex decision-making.
1 code implementation • 24 May 2023 • Tingting Ma, Qianhui Wu, Huiqiang Jiang, Börje F. Karlsson, Tiejun Zhao, Chin-Yew Lin
Cross-lingual named entity recognition (NER) aims to train an NER system that generalizes well to a target language by leveraging labeled data in a given source language.
no code implementations • 21 Jun 2023 • Mohan Zhou, Yalong Bai, Wei zhang, Ting Yao, Tiejun Zhao, Tao Mei
Dynamically synthesizing talking speech that actively responds to a listening head is critical during the face-to-face interaction.
no code implementations • 5 Jul 2023 • Mohan Zhou, Yalong Bai, Wei zhang, Ting Yao, Tiejun Zhao
Based on ViCo and ViCo-X, we define three novel tasks targeting the interaction modeling during the face-to-face conversation: 1) responsive listening head generation making listeners respond actively to the speaker with non-verbal signals, 2) expressive talking head generation guiding speakers to be aware of listeners' behaviors, and 3) conversational head generation to integrate the talking/listening ability in one interlocutor.
no code implementations • 20 Jul 2023 • Mohan Zhou, Yalong Bai, Wei zhang, Ting Yao, Tiejun Zhao, Tao Mei
In this paper, we propose a novel learning-based evaluation metric named Preference Score (PS) for fitting human preference according to the quantitative evaluations across different dimensions.
no code implementations • 22 Aug 2023 • Yingyao Wang, Yongwei Zhou, Chaoqun Duan, Junwei Bao, Tiejun Zhao
To alleviate these challenges, we propose a self-iterative framework for multi-hop program generation (HopPG) over heterogeneous knowledge, which leverages the previous execution results to retrieve supporting facts and generate subsequent programs hop by hop.
1 code implementation • 29 Jan 2024 • Jieru Lin, Danqing Huang, Tiejun Zhao, Dechen Zhan, Chin-Yew Lin
Furthermore, based on our observation that pixel space is more sensitive in capturing spatial patterns of graphic layouts (e. g., overlap, alignment), we propose a learning-based locator to detect erroneous tokens which takes the wireframe image rendered from the generated layout sequence as input.
no code implementations • 12 Feb 2024 • Hongyun Zhou, Xiangyu Lu, Wang Xu, Conghui Zhu, Tiejun Zhao
Low-Rank Adaptation (LoRA) introduces auxiliary parameters for each layer to fine-tune the pre-trained model under limited computing resources.
no code implementations • 5 Mar 2024 • Hui Huang, Yingqi Qu, Jing Liu, Muyun Yang, Tiejun Zhao
Recently, there has been a growing trend of utilizing Large Language Model (LLM) to evaluate the quality of other LLMs.
no code implementations • 7 Mar 2024 • Hui Huang, Yingqi Qu, Jing Liu, Muyun Yang, Tiejun Zhao
The proliferation of open-source Large Language Models (LLMs) underscores the pressing need for evaluation methods.
no code implementations • 27 Mar 2024 • Yongwei Zhou, Tiejun Zhao
To alleviate this problem, we propose a dual instruction tuning strategy to meticulously model mathematical reasoning from both forward and reverse directions.
no code implementations • CCL 2020 • Ting Jiang, Bing Xu, Tiejun Zhao, Sheng Li
In the first layer, in order to extract textual features of utterances, we propose a convolutional self-attention network(CAN).
no code implementations • Findings (NAACL) 2022 • Wang Xu, Tiejun Zhao
Abstractive summarization can generate high quality results with the development of the neural network.
no code implementations • COLING 2022 • Zihao Feng, Hailong Cao, Tiejun Zhao, Weixuan Wang, Wei Peng
Despite their progress in high-resource language settings, unsupervised bilingual lexicon induction (UBLI) models often fail on corpora with low-resource distant language pairs due to insufficient initialization.
1 code implementation • NAACL 2022 • Tingting Ma, Qianhui Wu, Zhiwei Yu, Tiejun Zhao, Chin-Yew Lin
Recent studies on few-shot intent detection have attempted to formulate the task as a meta-learning problem, where a meta-learning model is trained with a certain capability to quickly adapt to newly specified few-shot tasks with potentially unseen intent categories.
1 code implementation • NAACL 2022 • Yongwei Zhou, Junwei Bao, Chaoqun Duan, Haipeng Sun, Jiahui Liang, Yifan Wang, Jing Zhao, Youzheng Wu, Xiaodong He, Tiejun Zhao
To inherit the advantages of these two types of methods, we propose OPERA, an operation-pivoted discrete reasoning framework, where lightweight symbolic operations (compared with logical forms) as neural modules are utilized to facilitate the reasoning ability and interpretability.
no code implementations • CCL 2022 • Jiarui Tang, Liu Meiling, Tiejun Zhao, Jiyun Zhou
“神经网络模型的快速发展使得多文档摘要可以获得人类可读的流畅的摘要, 对大规模的数据进行预训练可以更好的从自然语言文本中捕捉更丰富的语义信息, 并更好的作用于下游任务。目前很多的多文档摘要的工作也应用了预训练模型(如BERT)并取得了一定的效果, 但是这些预训练模型不能更好的从文本中捕获事实性知识, 没有考虑到多文档文本的结构化的实体-关系信息, 本文提出了基于实体信息增强和多粒度融合的多文档摘要模型MGNIE, 将实体关系信息融入预训练模型ERNIE中, 增强知识事实以获得多层语义信息, 解决摘要生成的事实一致性问题。进而从多种粒度进行多文档层次结构的融合建模, 以词信息、实体信息以及句子信息捕捉长文本信息摘要生成所需的关键信息点。本文设计的模型, 在国际标准评测数据集MultiNews上对比强基线模型效果和竞争力获得较大提升。”