Author page Dong Yu (Idaho vs BLCU)

data/xml/2007.sigdial.xml

@@ -209,7 +209,7 @@
       <author><first>Ivan</first><last>Tashev</last></author>
       <author><first>Michael</first><last>Seltzer</last></author>
       <author><first>Yun-Cheng</first><last>Ju</last></author>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-idaho"><first>Dong</first><last>Yu</last></author>
       <author><first>Alex</first><last>Acero</last></author>
       <pages>87–94</pages>
       <url hash="f9be08f8">2007.sigdial-1.18</url>

data/xml/2012.iwslt.xml

@@ -22,7 +22,7 @@
     </paper>
     <paper id="3">
       <title>Who can understand your speech better – deep neural network of <fixed-case>G</fixed-case>aussian mixture model</title>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-idaho"><first>Dong</first><last>Yu</last></author>
       <url hash="f54a3245">2012.iwslt-keynotes.3</url>
       <bibkey>yu-2012-understand</bibkey>
     </paper>

data/xml/2020.acl.xml

@@ -1376,7 +1376,7 @@
       <author><first>Zhenyi</first><last>Wang</last></author>
       <author><first>Xiaoyang</first><last>Wang</last></author>
       <author><first>Bang</first><last>An</last></author>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-idaho"><first>Dong</first><last>Yu</last></author>
       <author><first>Changyou</first><last>Chen</last></author>
       <pages>1072–1086</pages>
       <abstract>Text generation from a knowledge base aims to translate knowledge triples to natural language descriptions. Most existing methods ignore the faithfulness between a generated text description and the original table, leading to generated information that goes beyond the content of the table. In this paper, for the first time, we propose a novel Transformer-based generation framework to achieve the goal. The core techniques in our method to enforce faithfulness include a new table-text optimal-transport matching loss and a table-text embedding similarity loss based on the Transformer model. Furthermore, to evaluate faithfulness, we propose a new automatic metric specialized to the table-to-text generation problem. We also provide detailed analysis on each component of our model in our experiments. Automatic and human evaluations show that our framework can significantly outperform state-of-the-art by a large margin.</abstract>
@@ -3140,7 +3140,7 @@
       <author><first>Jie</first><last>Lei</last></author>
       <author><first>Liwei</first><last>Wang</last></author>
       <author><first>Yelong</first><last>Shen</last></author>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-idaho"><first>Dong</first><last>Yu</last></author>
       <author><first>Tamara</first><last>Berg</last></author>
       <author><first>Mohit</first><last>Bansal</last></author>
       <pages>2603–2614</pages>
@@ -5985,7 +5985,7 @@
       <author><first>Dian</first><last>Yu</last></author>
       <author><first>Kai</first><last>Sun</last></author>
       <author><first>Claire</first><last>Cardie</last></author>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-idaho"><first>Dong</first><last>Yu</last></author>
       <pages>4927–4940</pages>
       <abstract>We present the first human-annotated dialogue-based relation extraction (RE) dataset DialogRE, aiming to support the prediction of relation(s) between two arguments that appear in a dialogue. We further offer DialogRE as a platform for studying cross-sentence RE as most facts span multiple sentences. We argue that speaker-related information plays a critical role in the proposed task, based on an analysis of similarities and differences between dialogue-based and traditional RE tasks. Considering the timeliness of communication in a dialogue, we design a new metric to evaluate the performance of RE methods in a conversational setting and investigate the performance of several representative RE methods on DialogRE. Experimental results demonstrate that a speaker-aware extension on the best-performing model leads to gains in both the standard and conversational evaluation settings. DialogRE is available at <url>https://dataset.org/dialogre/</url>.</abstract>
       <url hash="2c79b40c">2020.acl-main.444</url>
@@ -6482,7 +6482,7 @@
       <author><first>Kun</first><last>Xu</last></author>
       <author><first>Yue</first><last>Zhang</last></author>
       <author><first>Jianshu</first><last>Chen</last></author>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-idaho"><first>Dong</first><last>Yu</last></author>
       <pages>5429–5434</pages>
       <abstract>Zero pronoun recovery and resolution aim at recovering the dropped pronoun and pointing out its anaphoric mentions, respectively. We propose to better explore their interaction by solving both tasks together, while the previous work treats them separately. For zero pronoun resolution, we study this task in a more realistic setting, where no parsing trees or only automatic trees are available, while most previous work assumes gold trees. Experiments on two benchmarks show that joint modeling significantly outperforms our baseline that already beats the previous state of the arts.</abstract>
       <url hash="5af6f516">2020.acl-main.482</url>
@@ -8139,7 +8139,7 @@
       <author><first>Yelong</first><last>Shen</last></author>
       <author><first>Dian</first><last>Yu</last></author>
       <author><first>Jianshu</first><last>Chen</last></author>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-idaho"><first>Dong</first><last>Yu</last></author>
       <pages>6751–6761</pages>
       <abstract>In this paper, we study machine reading comprehension (MRC) on long texts: where a model takes as inputs a lengthy document and a query, extracts a text span from the document as an answer. State-of-the-art models (e.g., BERT) tend to use a stack of transformer layers that are pre-trained from a large number of unlabeled language corpora to encode the joint contextual information of query and document. However, these transformer models can only take as input a fixed-length (e.g., 512) text. To deal with even longer text inputs, previous approaches usually chunk them into <i>equally-spaced</i> segments and predict answers based on each segment independently without considering the information from other segments. As a result, they may form segments that fail to cover complete answers or retain insufficient contexts around the correct answer required for question answering. Moreover, they are less capable of answering questions that need cross-segment information. We propose to let a model learn to chunk in a more flexible way via reinforcement learning: a model can decide the next segment that it wants to process in either direction. We also apply recurrent mechanisms to enable information to flow across segments. Experiments on three MRC tasks – CoQA, QuAC, and TriviaQA – demonstrate the effectiveness of our proposed recurrent chunking mechanisms: we can obtain segments that are more likely to contain complete answers and at the same time provide sufficient contexts around the ground truth answers for better predictions.</abstract>
       <url hash="ffc15051">2020.acl-main.603</url>
@@ -9580,7 +9580,7 @@
       <author><first>Yue</first><last>Zhang</last></author>
       <author><first>Kun</first><last>Xu</last></author>
       <author><first>Yubin</first><last>Ge</last></author>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-idaho"><first>Dong</first><last>Yu</last></author>
       <pages>7987–7998</pages>
       <abstract>The task of graph-to-text generation aims at producing sentences that preserve the meaning of input graphs. As a crucial defect, the current state-of-the-art models may mess up or even drop the core structural information of input graphs when generating outputs. We propose to tackle this problem by leveraging richer training signals that can guide our model for preserving input information. In particular, we introduce two types of autoencoding losses, each individually focusing on different aspects (a.k.a. views) of input graphs. The losses are then back-propagated to better calibrate our model via multi-task training. Experiments on two benchmarks for graph-to-text generation show the effectiveness of our approach over a state-of-the-art baseline.</abstract>
       <url hash="c39534ed">2020.acl-main.712</url>

data/xml/2020.ccl.xml

@@ -590,7 +590,7 @@
       <title>面向人工智能伦理计算的中文道德词典构建方法研究(Construction of a <fixed-case>C</fixed-case>hinese Moral Dictionary for Artificial Intelligence Ethical Computing)</title>
       <author><first>Hongrui</first><last>Wang</last><variant script="hani"><first>弘睿</first><last>王</last></variant></author>
       <author><first>Chang</first><last>Liu</last><variant script="hani"><first>畅</first><last>刘</last></variant></author>
-      <author><first>Dong</first><last>Yu</last><variant script="hani"><first>东</first><last>于</last></variant></author>
+      <author id="dong-yu-blcu"><first>Dong</first><last>Yu</last><variant script="hani"><first>东</first><last>于</last></variant></author>
       <pages>539–549</pages>
       <abstract>道德词典资源的建设是人工智能伦理计算的一个研究重点。由于道德行为复杂多样,现有的英文道德词典分类体系并不完善,而中文方面目前尚未有相关的词典资源,理论体系和构建方法仍待探究。针对以上问题,该文提出了面向人工智能伦理计算的中文道德词典构建任务,设计了四类标签和四种类型,得到包含25,012个词的中文道德词典资源。实验结果表明,该词典资源不仅能够使机器学会道德知识,判断词的道德标签和类型,而且能够为句子级别的道德文本分析提供数据支持。</abstract>
       <url hash="c5445fa1">2020.ccl-1.50</url>
@@ -811,7 +811,7 @@
     <paper id="68">
       <title>结合深度学习和语言难度特征的句子可读性计算方法(The method of calculating sentence readability combined with deep learning and language difficulty characteristics)</title>
       <author><first>Yuling</first><last>Tang</last><variant script="hani"><first>玉玲</first><last>唐</last></variant></author>
-      <author><first>Dong</first><last>Yu</last><variant script="hani"><first>东</first><last>于</last></variant></author>
+      <author id="dong-yu-blcu"><first>Dong</first><last>Yu</last><variant script="hani"><first>东</first><last>于</last></variant></author>
       <pages>731–742</pages>
       <abstract>本文提出了可读性语料库构建的改进方法,基于该方法,构建了规模更大的汉语句子可读性语料库。该语料库在句子绝对难度评估任务上的准确率达到0.7869,相对前人工作提升了0.15以上,证明了改进方法的有效性。将深度学习方法应用于汉语可读性评估,探究了不同深度学习方法自动捕获难度特征的能力,并进仛步探究了向深度学习特征中融入不同层面的语难度特征对模型整体性能的影响。实验结果显示,不同深度学习模型的难度特征捕获能力不尽相同,语言难度特征可以不同程度地提高深度学习模型的难度表征能力。</abstract>
       <url hash="99223a1a">2020.ccl-1.68</url>

data/xml/2020.emnlp.xml

@@ -1030,7 +1030,7 @@
       <author><first>Yang</first><last>Feng</last></author>
       <author><first>Wanying</first><last>Xie</last></author>
       <author><first>Jie</first><last>Zhou</last></author>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-blcu"><first>Dong</first><last>Yu</last></author>
       <pages>1035–1046</pages>
       <abstract>There exists a token imbalance phenomenon in natural language as different tokens appear with different frequencies, which leads to different learning difficulties for tokens in Neural Machine Translation (NMT). The vanilla NMT model usually adopts trivial equal-weighted objectives for target tokens with different frequencies and tends to generate more high-frequency tokens and less low-frequency tokens compared with the golden token distribution. However, low-frequency tokens may carry critical semantic information that will affect the translation quality once they are neglected. In this paper, we explored target token-level adaptive objectives based on token frequencies to assign appropriate weights for each target token during training. We aimed that those meaningful but relatively low-frequency words could be assigned with larger weights in objectives to encourage the model to pay more attention to these tokens. Our method yields consistent improvements in translation quality on ZH-EN, EN-RO, and EN-DE translation tasks, especially on sentences that contain more low-frequency tokens where we can get 1.68, 1.02, and 0.52 BLEU increases compared with baseline, respectively. Further analyses show that our method can also improve the lexical diversity of translation.</abstract>
       <url hash="3986c73c">2020.emnlp-main.76</url>
@@ -6836,7 +6836,7 @@
       <author><first>Sangwoo</first><last>Cho</last></author>
       <author><first>Kaiqiang</first><last>Song</last></author>
       <author><first>Chen</first><last>Li</last></author>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-idaho"><first>Dong</first><last>Yu</last></author>
       <author><first>Hassan</first><last>Foroosh</last></author>
       <author id="fei-liu-utdallas"><first>Fei</first><last>Liu</last></author>
       <pages>6282–6300</pages>
@@ -7214,7 +7214,7 @@
       <author><first>Han</first><last>Wu</last></author>
       <author><first>Haisong</first><last>Zhang</last></author>
       <author><first>Linqi</first><last>Song</last></author>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-idaho"><first>Dong</first><last>Yu</last></author>
       <pages>6632–6639</pages>
       <abstract>For multi-turn dialogue rewriting, the capacity of effectively modeling the linguistic knowledge in dialog context and getting ride of the noises is essential to improve its performance. Existing attentive models attend to all words without prior focus, which results in inaccurate concentration on some dispensable words. In this paper, we propose to use semantic role labeling (SRL), which highlights the core semantic information of who did what to whom, to provide additional guidance for the rewriter model. Experiments show that this information significantly improves a RoBERTa-based model that already outperforms previous state-of-the-art systems.</abstract>
       <url hash="34118239">2020.emnlp-main.537</url>

data/xml/2020.semeval.xml

@@ -354,7 +354,7 @@
       <author><first>Shike</first><last>Wang</last></author>
       <author><first>Yuchen</first><last>Fan</last></author>
       <author><first>Xiangying</first><last>Luo</last></author>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-blcu"><first>Dong</first><last>Yu</last></author>
       <pages>255–262</pages>
       <abstract>Lexical entailment recognition plays an important role in tasks like Question Answering and Machine Translation. As important branches of lexical entailment, predicting multilingual and cross-lingual lexical entailment (LE) are two subtasks of SemEval2020 Task2. In previous monolingual LE studies, researchers leverage external linguistic constraints to transform word embeddings for LE relation. In our system, we expand the number of external constraints in multiple languages to obtain more specialised multilingual word embeddings. For the cross-lingual subtask, we apply a bilingual word embeddings mapping method in the model. The mapping method takes specialised embeddings as inputs and is able to retain the embeddings’ LE features after operations. Our results for multilingual subtask are about 20% and 10% higher than the baseline in graded and binary prediction respectively.</abstract>
       <url hash="0c4cb83a">2020.semeval-1.31</url>
@@ -930,7 +930,7 @@
     <paper id="81">
       <title><fixed-case>BLCU</fixed-case>-<fixed-case>NLP</fixed-case> at <fixed-case>S</fixed-case>em<fixed-case>E</fixed-case>val-2020 Task 5: Data Augmentation for Efficient Counterfactual Detecting</title>
       <author><first>Chang</first><last>Liu</last></author>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-blcu"><first>Dong</first><last>Yu</last></author>
       <pages>633–639</pages>
       <abstract>Counterfactuals describe events counter to facts and hence naturally involve common sense, knowledge, and reasoning. SemEval 2020 task 5 is focusing on this field. We participate in the subtask 1 and we use BERT as our system. Our Innovations are feature extraction and data augmentation. We extract and summarize features of counterfactual statements, augment counterfactual examples in training set with the help of these features, and two general methods of data augmentation is experimented in our work. We demonstrate the effectiveness of our approaches, which achieves 0.95 of subtask 1 in F1 while using only a subset of giving training set to fine-tune the BERT model, and our official submission achieves F1 0.802, which ranks us 16th in the competition.</abstract>
       <url hash="2d0e62d4">2020.semeval-1.81</url>

data/xml/2020.tacl.xml

@@ -123,7 +123,7 @@
       <title>Investigating Prior Knowledge for Challenging <fixed-case>C</fixed-case>hinese Machine Reading Comprehension</title>
       <author><first>Kai</first><last>Sun</last></author>
       <author><first>Dian</first><last>Yu</last></author>
-      <author><first>Dong</first><last>Yu</last></author>
+      <author id="dong-yu-idaho"><first>Dong</first><last>Yu</last></author>
       <author><first>Claire</first><last>Cardie</last></author>
       <doi>10.1162/tacl_a_00305</doi>
       <abstract>Machine reading comprehension tasks require a machine reader to answer questions relevant to the given document. In this paper, we present the first free-form multiple-Choice Chinese machine reading Comprehension dataset (C3), containing 13,369 documents (dialogues or more formally written mixed-genre texts) and their associated 19,577 multiple-choice free-form questions collected from Chinese-as-a-second-language examinations. We present a comprehensive analysis of the prior knowledge (i.e., linguistic, domain-specific, and general world knowledge) needed for these real-world problems. We implement rule-based and popular neural methods and find that there is still a significant performance gap between the best performing model (68.5%) and human readers (96.0%), especiallyon problems that require prior knowledge. We further study the effects of distractor plausibility and data augmentation based on translated relevant datasets for English on model performance. We expect C3 to present great challenges to existing systems as answering 86.8% of questions requires both knowledge within and beyond the accompanying document, and we hope that C3 can serve as a platform to study how to leverage various kinds of prior knowledge to better understand a given written or orally oriented text. C3 is available at <url>https://dataset.org/c3/</url>.</abstract>