October 16, 2019

2793 words 14 mins read

Paper Group NAWR 12

The effect of information controls on developers in China: An analysis of censorship in Chinese open source projects. Improving Hate Speech Detection with Deep Learning Ensembles. Automatic Dialogue Generation with Expressed Emotions. Adversarial Multi-lingual Neural Relation Extraction. Learning Loop Invariants for Program Verification. A Melody-C …

The effect of information controls on developers in China: An analysis of censorship in Chinese open source projects


Title	The effect of information controls on developers in China: An analysis of censorship in Chinese open source projects
Authors	Jeffrey Knockel, Masashi Crete-Nishihata, Lotus Ruan
Abstract	Censorship of Internet content in China is understood to operate through a system of intermediary liability whereby service providers are liable for the content on their platforms. Previous work studying censorship has found huge variability in the implementation of censorship across different products even within the same industry segment. In this work we explore the extent to which these censorship features are present in the open source projects of individual developers in China by collecting their blacklists and comparing their similarity. We collect files from a popular online code repository, extract lists of strings, and then classify whether each is a Chinese blacklist. Overall, we found over 1,000 Chinese blacklists comprising over 200,000 unique keywords, representing the largest dataset of Chinese blacklisted keywords to date. We found very little keyword overlap between lists, raising questions as to their origins, as the lists seem too large to have been individually curated, yet the lack of overlap suggests that they have no common source.
Tasks
Published	2018-08-01
URL	https://www.aclweb.org/anthology/W18-4201/
PDF	https://www.aclweb.org/anthology/W18-4201
PWC	https://paperswithcode.com/paper/the-effect-of-information-controls-on
Repo	https://github.com/citizenlab/chat-censorship
Framework	none

Improving Hate Speech Detection with Deep Learning Ensembles


Title	Improving Hate Speech Detection with Deep Learning Ensembles
Authors	Steven Zimmerman, Udo Kruschwitz, Chris Fox
Abstract
Tasks	Hate Speech Detection, Sentiment Analysis, Text Classification
Published	2018-05-01
URL	https://www.aclweb.org/anthology/L18-1404/
PDF	https://www.aclweb.org/anthology/L18-1404
PWC	https://paperswithcode.com/paper/improving-hate-speech-detection-with-deep
Repo	https://github.com/stevenzim/lrec-2018
Framework	none

Automatic Dialogue Generation with Expressed Emotions


Title	Automatic Dialogue Generation with Expressed Emotions
Authors	Chenyang Huang, Osmar Za{"\i}ane, Amine Trabelsi, Nouha Dziri
Abstract	Despite myriad efforts in the literature designing neural dialogue generation systems in recent years, very few consider putting restrictions on the response itself. They learn from collections of past responses and generate one based on a given utterance without considering, speech act, desired style or emotion to be expressed. In this research, we address the problem of forcing the dialogue generation to express emotion. We present three models that either concatenate the desired emotion with the source input during the learning, or push the emotion in the decoder. The results, evaluated with an emotion tagger, are encouraging with all three models, but present better outcome and promise with our model that adds the emotion vector in the decoder.
Tasks	Dialogue Generation
Published	2018-06-01
URL	https://www.aclweb.org/anthology/N18-2008/
PDF	https://www.aclweb.org/anthology/N18-2008
PWC	https://paperswithcode.com/paper/automatic-dialogue-generation-with-expressed
Repo	https://github.com/chenyangh/DialogueGenerationWithEmotion
Framework	pytorch

Adversarial Multi-lingual Neural Relation Extraction


Title	Adversarial Multi-lingual Neural Relation Extraction
Authors	Xiaozhi Wang, Xu Han, Yankai Lin, Zhiyuan Liu, Maosong Sun
Abstract	Multi-lingual relation extraction aims to find unknown relational facts from text in various languages. Existing models cannot well capture the consistency and diversity of relation patterns in different languages. To address these issues, we propose an adversarial multi-lingual neural relation extraction (AMNRE) model, which builds both consistent and individual representations for each sentence to consider the consistency and diversity among languages. Further, we adopt an adversarial training strategy to ensure those consistent sentence representations could effectively extract the language-consistent relation patterns. The experimental results on real-world datasets demonstrate that our AMNRE model significantly outperforms the state-of-the-art models. The source code of this paper can be obtained from https://github.com/thunlp/AMNRE.
Tasks	Question Answering, Relation Extraction
Published	2018-08-01
URL	https://www.aclweb.org/anthology/C18-1099/
PDF	https://www.aclweb.org/anthology/C18-1099
PWC	https://paperswithcode.com/paper/adversarial-multi-lingual-neural-relation
Repo	https://github.com/thunlp/AMNRE
Framework	pytorch

Learning Loop Invariants for Program Verification


Title	Learning Loop Invariants for Program Verification
Authors	Xujie Si, Hanjun Dai, Mukund Raghothaman, Mayur Naik, Le Song
Abstract	A fundamental problem in program verification concerns inferring loop invariants. The problem is undecidable and even practical instances are challenging. Inspired by how human experts construct loop invariants, we propose a reasoning framework Code2Inv that constructs the solution by multi-step decision making and querying an external program graph memory block. By training with reinforcement learning, Code2Inv captures rich program features and avoids the need for ground truth solutions as supervision. Compared to previous learning tasks in domains with graph-structured data, it addresses unique challenges, such as a binary objective function and an extremely sparse reward that is given by an automated theorem prover only after the complete loop invariant is proposed. We evaluate Code2Inv on a suite of 133 benchmark problems and compare it to three state-of-the-art systems. It solves 106 problems compared to 73 by a stochastic search-based system, 77 by a heuristic search-based system, and 100 by a decision tree learning-based system. Moreover, the strategy learned can be generalized to new programs: compared to solving new instances from scratch, the pre-trained agent is more sample efficient in finding solutions.
Tasks	Decision Making
Published	2018-12-01
URL	http://papers.nips.cc/paper/8001-learning-loop-invariants-for-program-verification
PDF	http://papers.nips.cc/paper/8001-learning-loop-invariants-for-program-verification.pdf
PWC	https://paperswithcode.com/paper/learning-loop-invariants-for-program
Repo	https://github.com/PL-ML/code2inv
Framework	pytorch

A Melody-Conditioned Lyrics Language Model


Title	A Melody-Conditioned Lyrics Language Model
Authors	Kento Watanabe, Yuichiroh Matsubayashi, Satoru Fukayama, Masataka Goto, Kentaro Inui, Tomoyasu Nakano
Abstract	This paper presents a novel, data-driven language model that produces entire lyrics for a given input melody. Previously proposed models for lyrics generation suffer from the inability of capturing the relationship between lyrics and melody partly due to the unavailability of lyrics-melody aligned data. In this study, we first propose a new practical method for creating a large collection of lyrics-melody aligned data and then create a collection of 1,000 lyrics-melody pairs augmented with precise syllable-note alignments and word/sentence/paragraph boundaries. We then provide a quantitative analysis of the correlation between word/sentence/paragraph boundaries in lyrics and melodies. We then propose an RNN-based lyrics language model conditioned on a featurized melody. Experimental results show that the proposed model generates fluent lyrics while maintaining the compatibility between boundaries of lyrics and melody structures.
Tasks	Language Modelling
Published	2018-06-01
URL	https://www.aclweb.org/anthology/N18-1015/
PDF	https://www.aclweb.org/anthology/N18-1015
PWC	https://paperswithcode.com/paper/a-melody-conditioned-lyrics-language-model
Repo	https://github.com/KentoW/melody-lyrics
Framework	none

Fast Sequence Based Embedding with Diffusion Graphs


Title	Fast Sequence Based Embedding with Diffusion Graphs
Authors	Benedek Rozemberczki, Rik Sarkar
Abstract	A graph embedding is a representation of the vertices of a graph in a low dimensional space, which approximately preserves proper-ties such as distances between nodes. Vertex sequence based embedding procedures use features extracted from linear sequences of vertices to create embeddings using a neural network. In this paper, we propose diffusion graphs as a method to rapidly generate vertex sequences for network embedding. Its computational efficiency is superior to previous methods due to simpler sequence generation, and it produces more ac-curate results. In experiments, we found that the performance relative to other methods improves with increasing edge density in the graph.In a community detection task, clustering nodes in the embedding space produces better results compared to other sequence based embedding methods.
Tasks	Community Detection, Graph Embedding, Network Embedding, Node Classification
Published	2018-03-20
URL	http://homepages.inf.ed.ac.uk/s1668259/papers/sequence.pdf
PDF	http://homepages.inf.ed.ac.uk/s1668259/papers/sequence.pdf
PWC	https://paperswithcode.com/paper/fast-sequence-based-embedding-with-diffusion
Repo	https://github.com/benedekrozemberczki/karateclub
Framework	none

Probabilistic Neural Programmed Networks for Scene Generation


Title	Probabilistic Neural Programmed Networks for Scene Generation
Authors	Zhiwei Deng, Jiacheng Chen, Yifang Fu, Greg Mori
Abstract	In this paper we address the text to scene image generation problem. Generative models that capture the variability in complicated scenes containing rich semantics is a grand goal of image generation. Complicated scene images contain rich visual elements, compositional visual concepts, and complicated relations between objects. Generative models, as an analysis-by-synthesis process, should encompass the following three core components: 1) the generation process that composes the scene; 2) what are the primitive visual elements and how are they composed; 3) the rendering of abstract concepts into their pixel-level realizations. We propose PNP-Net, a variational auto-encoder framework that addresses these three challenges: it flexibly composes images with a dynamic network structure, learns a set of distribution transformers that can compose distributions based on semantics, and decodes samples from these distributions into realistic images.
Tasks	Image Generation, Scene Generation
Published	2018-12-01
URL	http://papers.nips.cc/paper/7658-probabilistic-neural-programmed-networks-for-scene-generation
PDF	http://papers.nips.cc/paper/7658-probabilistic-neural-programmed-networks-for-scene-generation.pdf
PWC	https://paperswithcode.com/paper/probabilistic-neural-programmed-networks-for
Repo	https://github.com/Lucas2012/ProbabilisticNeuralProgrammedNetwork
Framework	pytorch

Attention-based Multi-Patch Aggregation for Image Aesthetic Assessment


Title	Attention-based Multi-Patch Aggregation for Image Aesthetic Assessment
Authors	Kekai Sheng, Weiming Dong, Chongyang Ma, Xing Mei, Feiyue Huang, Bao-Gang Hu
Abstract	Aggregation structures with explicit information, such as image attributes and scene semantics, are effective and popular for intelligent systems for assessing aesthetics of visual data. However, useful information may not be available due to the high cost of manual annotation and expert design. In this paper, we present a novel multi-patch (MP) aggregation method for image aesthetic assessment. Different from state-of-the-art methods, which augment an MP aggregation network with various visual attributes, we train the model in an end-to-end manner with aesthetic labels only (i.e., aesthetically positive or negative). We achieve the goal by resorting to an attention-based mechanism that adaptively adjusts the weight of each patch during the training process to improve learning efficiency. In addition, we propose a set of objectives with three typical attention mechanisms (i.e., average, minimum, and adaptive) and evaluate their effectiveness on the Aesthetic Visual Analysis (AVA) benchmark. Numerical results show that our approach outperforms existing methods by a large margin. We further verify the effectiveness of the proposed attention-based objectives via ablation studies and shed light on the design of aesthetic assessment systems.
Tasks	Aesthetics Quality Assessment
Published	2018-10-22
URL	https://www.researchgate.net/publication/328371233_Attention-based_Multi-Patch_Aggregation_for_Image_Aesthetic_Assessment
PDF	https://www.researchgate.net/publication/328371233_Attention-based_Multi-Patch_Aggregation_for_Image_Aesthetic_Assessment
PWC	https://paperswithcode.com/paper/attention-based-multi-patch-aggregation-for
Repo	https://github.com/Openning07/MPADA
Framework	tf

RT-GENE: Real-Time Eye Gaze Estimation in Natural Environments


Title	RT-GENE: Real-Time Eye Gaze Estimation in Natural Environments
Authors	Tobias Fischer, Hyung Jin Chang, Yiannis Demiris
Abstract	In this work, we consider the problem of robust gaze estimation in natural environments. Large camera-to-subject distances and high variations in head pose and eye gaze angles are common in such environments. This leads to two main shortfalls in state-of-the-art methods for gaze estimation: hindered ground truth gaze annotation and diminished gaze estimation accuracy as image resolution decreases with distance. We first record a novel dataset of varied gaze and head pose images in a natural environment, addressing the issue of ground truth annotation by measuring head pose using a motion capture system and eye gaze using mobile eyetracking glasses. We apply semantic image inpainting to the area covered by the glasses to bridge the gap between training and testing images by removing the obtrusiveness of the glasses. We also present a new real-time algorithm involving appearance-based deep convolutional neural networks with increased capacity to cope with the diverse images in the new dataset. Experiments with this network architecture are conducted on a number of diverse eye-gaze datasets including our own, and in cross dataset evaluations. We demonstrate state-of-the-art performance in terms of estimation accuracy in all experiments, and the architecture performs well even on lower resolution images.
Tasks	Gaze Estimation, Image Inpainting, Motion Capture
Published	2018-10-06
URL	http://openaccess.thecvf.com/content_ECCV_2018/html/Tobias_Fischer_RT-GENE_Real-Time_Eye_ECCV_2018_paper.html
PDF	http://openaccess.thecvf.com/content_ECCV_2018/papers/Tobias_Fischer_RT-GENE_Real-Time_Eye_ECCV_2018_paper.pdf
PWC	https://paperswithcode.com/paper/rt-gene-real-time-eye-gaze-estimation-in
Repo	https://github.com/Tobias-Fischer/rt_gene
Framework	tf

Representation Learning for Treatment Effect Estimation from Observational Data


Title	Representation Learning for Treatment Effect Estimation from Observational Data
Authors	Liuyi Yao, Sheng Li, Yaliang Li, Mengdi Huai, Jing Gao, Aidong Zhang
Abstract	Estimating individual treatment effect (ITE) is a challenging problem in causal inference, due to the missing counterfactuals and the selection bias. Existing ITE estimation methods mainly focus on balancing the distributions of control and treated groups, but ignore the local similarity information that is helpful. In this paper, we propose a local similarity preserved individual treatment effect (SITE) estimation method based on deep representation learning. SITE preserves local similarity and balances data distributions simultaneously, by focusing on several hard samples in each mini-batch. Experimental results on synthetic and three real-world datasets demonstrate the advantages of the proposed SITE method, compared with the state-of-the-art ITE estimation methods.
Tasks	Causal Inference, Representation Learning
Published	2018-12-01
URL	http://papers.nips.cc/paper/7529-representation-learning-for-treatment-effect-estimation-from-observational-data
PDF	http://papers.nips.cc/paper/7529-representation-learning-for-treatment-effect-estimation-from-observational-data.pdf
PWC	https://paperswithcode.com/paper/representation-learning-for-treatment-effect
Repo	https://github.com/Osier-Yi/SITE
Framework	tf

SpatialVOC2K: A Multilingual Dataset of Images with Annotations and Features for Spatial Relations between Objects


Title	SpatialVOC2K: A Multilingual Dataset of Images with Annotations and Features for Spatial Relations between Objects
Authors	Anja Belz, Adrian Muscat, Pierre Anguill, Mouhamadou Sow, Ga{'e}tan Vincent, Yassine Zinessabah
Abstract	We present SpatialVOC2K, the first multilingual image dataset with spatial relation annotations and object features for image-to-text generation, built using 2,026 images from the PASCAL VOC2008 dataset. The dataset incorporates (i) the labelled object bounding boxes from VOC2008, (ii) geometrical, language and depth features for each object, and (iii) for each pair of objects in both orders, (a) the single best preposition and (b) the set of possible prepositions in the given language that describe the spatial relationship between the two objects. Compared to previous versions of the dataset, we have roughly doubled the size for French, and completely reannotated as well as increased the size of the English portion, providing single best prepositions for English for the first time. Furthermore, we have added explicit 3D depth features for objects. We are releasing our dataset for free reuse, along with evaluation tools to enable comparative evaluation.
Tasks	Text Generation
Published	2018-11-01
URL	https://www.aclweb.org/anthology/W18-6516/
PDF	https://www.aclweb.org/anthology/W18-6516
PWC	https://paperswithcode.com/paper/spatialvoc2k-a-multilingual-dataset-of-images
Repo	https://github.com/muskata/SpatialVOC2K
Framework	none

To Attend or not to Attend: A Case Study on Syntactic Structures for Semantic Relatedness


Title	To Attend or not to Attend: A Case Study on Syntactic Structures for Semantic Relatedness
Authors	Amulya Gupta, Zhu Zhang
Abstract	With the recent success of Recurrent Neural Networks (RNNs) in Machine Translation (MT), attention mechanisms have become increasingly popular. The purpose of this paper is two-fold; firstly, we propose a novel attention model on Tree Long Short-Term Memory Networks (Tree-LSTMs), a tree-structured generalization of standard LSTM. Secondly, we study the interaction between attention and syntactic structures, by experimenting with three LSTM variants: bidirectional-LSTMs, Constituency Tree-LSTMs, and Dependency Tree-LSTMs. Our models are evaluated on two semantic relatedness tasks: semantic relatedness scoring for sentence pairs (SemEval 2012, Task 6 and SemEval 2014, Task 1) and paraphrase detection for question pairs (Quora, 2017).
Tasks	Machine Translation, Paraphrase Identification, Question Answering
Published	2018-07-01
URL	https://www.aclweb.org/anthology/P18-1197/
PDF	https://www.aclweb.org/anthology/P18-1197
PWC	https://paperswithcode.com/paper/to-attend-or-not-to-attend-a-case-study-on
Repo	https://github.com/amulyahwr/acl2018
Framework	pytorch

Incorporating Latent Meanings of Morphological Compositions to Enhance Word Embeddings


Title	Incorporating Latent Meanings of Morphological Compositions to Enhance Word Embeddings
Authors	Yang Xu, Jiawei Liu, Wei Yang, Liusheng Huang
Abstract	Traditional word embedding approaches learn semantic information at word level while ignoring the meaningful internal structures of words like morphemes. Furthermore, existing morphology-based models directly incorporate morphemes to train word embeddings, but still neglect the latent meanings of morphemes. In this paper, we explore to employ the latent meanings of morphological compositions of words to train and enhance word embeddings. Based on this purpose, we propose three Latent Meaning Models (LMMs), named LMM-A, LMM-S and LMM-M respectively, which adopt different strategies to incorporate the latent meanings of morphemes during the training process. Experiments on word similarity, syntactic analogy and text classification are conducted to validate the feasibility of our models. The results demonstrate that our models outperform the baselines on five word similarity datasets. On Wordsim-353 and RG-65 datasets, our models nearly achieve 5{%} and 7{%} gains over the classic CBOW model, respectively. For the syntactic analogy and text classification tasks, our models also surpass all the baselines including a morphology-based model.
Tasks	Information Retrieval, Machine Translation, Morphological Analysis, Sentiment Analysis, Text Classification, Word Embeddings
Published	2018-07-01
URL	https://www.aclweb.org/anthology/P18-1114/
PDF	https://www.aclweb.org/anthology/P18-1114
PWC	https://paperswithcode.com/paper/incorporating-latent-meanings-of
Repo	https://github.com/Y-Xu/lmm
Framework	none

From Strings to Other Things: Linking the Neighborhood and Transposition Effects in Word Reading


Title	From Strings to Other Things: Linking the Neighborhood and Transposition Effects in Word Reading
Authors	St{'e}phan Tulkens, S, Dominiek ra, Walter Daelemans
Abstract	We investigate the relation between the transposition and deletion effects in word reading, i.e., the finding that readers can successfully read {`}SLAT{''} as {`}SALT{''}, or {`}WRK{''} as {`}WORK{''}, and the neighborhood effect. In particular, we investigate whether lexical orthographic neighborhoods take into account transposition and deletion in determining neighbors. If this is the case, it is more likely that the neighborhood effect takes place early during processing, and does not solely rely on similarity of internal representations. We introduce a new neighborhood measure, rd20, which can be used to quantify neighborhood effects over arbitrary feature spaces. We calculate the rd20 over large sets of words in three languages using various feature sets and show that feature sets that do not allow for transposition or deletion explain more variance in Reaction Time (RT) measurements. We also show that the rd20 can be calculated using the hidden state representations of an Multi-Layer Perceptron, and show that these explain less variance than the raw features. We conclude that the neighborhood effect is unlikely to have a perceptual basis, but is more likely to be the result of items co-activating after recognition. All code is available at: \url{www.github.com/clips/conll2018}
Tasks
Published	2018-10-01
URL	https://www.aclweb.org/anthology/K18-1008/
PDF	https://www.aclweb.org/anthology/K18-1008
PWC	https://paperswithcode.com/paper/from-strings-to-other-things-linking-the
Repo	https://github.com/clips/conll2018
Framework	none