Paper Group NANR 200
Butterfly Effects in Frame Semantic Parsing: impact of data processing on model ranking. A Deep Learning Approach for Survival Clustering without End-of-life Signals. Gated Multi-Task Network for Text Classification. Universality, Robustness, and Detectability of Adversarial Perturbations under Adversarial Training. Development of a Mobile Observat …
Butterfly Effects in Frame Semantic Parsing: impact of data processing on model ranking
| Title | Butterfly Effects in Frame Semantic Parsing: impact of data processing on model ranking |
| Authors | Alex Kabbach, re, Corentin Ribeyre, Aur{'e}lie Herbelot |
| Abstract | Knowing the state-of-the-art for a particular task is an essential component of any computational linguistics investigation. But can we be truly confident that the current state-of-the-art is indeed the best performing model? In this paper, we study the case of frame semantic parsing, a well-established task with multiple shared datasets. We show that in spite of all the care taken to provide a standard evaluation resource, small variations in data processing can have dramatic consequences for ranking parser performance. This leads us to propose an open-source standardized processing pipeline, which can be shared and reused for robust model comparison. |
| Tasks | Semantic Parsing |
| Published | 2018-08-01 |
| URL | https://www.aclweb.org/anthology/C18-1267/ |
| https://www.aclweb.org/anthology/C18-1267 | |
| PWC | https://paperswithcode.com/paper/butterfly-effects-in-frame-semantic-parsing |
| Repo | |
| Framework | |
A Deep Learning Approach for Survival Clustering without End-of-life Signals
| Title | A Deep Learning Approach for Survival Clustering without End-of-life Signals |
| Authors | S Chandra Mouli, Bruno Ribeiro, Jennifer Neville |
| Abstract | The goal of survival clustering is to map subjects (e.g., users in a social network, patients in a medical study) to $K$ clusters ranging from low-risk to high-risk. Existing survival methods assume the presence of clear \textit{end-of-life} signals or introduce them artificially using a pre-defined timeout. In this paper, we forego this assumption and introduce a loss function that differentiates between the empirical lifetime distributions of the clusters using a modified Kuiper statistic. We learn a deep neural network by optimizing this loss, that performs a soft clustering of users into survival groups. We apply our method to a social network dataset with over 1M subjects, and show significant improvement in C-index compared to alternatives. |
| Tasks | |
| Published | 2018-01-01 |
| URL | https://openreview.net/forum?id=SJme6-ZR- |
| https://openreview.net/pdf?id=SJme6-ZR- | |
| PWC | https://paperswithcode.com/paper/a-deep-learning-approach-for-survival |
| Repo | |
| Framework | |
Gated Multi-Task Network for Text Classification
| Title | Gated Multi-Task Network for Text Classification |
| Authors | Liqiang Xiao, Honglun Zhang, Wenqing Chen |
| Abstract | Multi-task learning with Convolutional Neural Network (CNN) has shown great success in many Natural Language Processing (NLP) tasks. This success can be largely attributed to the feature sharing by fusing some layers among tasks. However, most existing approaches just fully or proportionally share the features without distinguishing the helpfulness of them. By that the network would be confused by the helpless even harmful features, generating undesired interference between tasks. In this paper, we introduce gate mechanism into multi-task CNN and propose a new Gated Sharing Unit, which can filter the feature flows between tasks and greatly reduce the interference. Experiments on 9 text classification datasets shows that our approach can learn selection rules automatically and gain a great improvement over strong baselines. |
| Tasks | Multi-Task Learning, Text Classification |
| Published | 2018-06-01 |
| URL | https://www.aclweb.org/anthology/N18-2114/ |
| https://www.aclweb.org/anthology/N18-2114 | |
| PWC | https://paperswithcode.com/paper/gated-multi-task-network-for-text |
| Repo | |
| Framework | |
Universality, Robustness, and Detectability of Adversarial Perturbations under Adversarial Training
| Title | Universality, Robustness, and Detectability of Adversarial Perturbations under Adversarial Training |
| Authors | Jan Hendrik Metzen |
| Abstract | Classifiers such as deep neural networks have been shown to be vulnerable against adversarial perturbations on problems with high-dimensional input space. While adversarial training improves the robustness of classifiers against such adversarial perturbations, it leaves classifiers sensitive to them on a non-negligible fraction of the inputs. We argue that there are two different kinds of adversarial perturbations: shared perturbations which fool a classifier on many inputs and singular perturbations which only fool the classifier on a small fraction of the data. We find that adversarial training increases the robustness of classifiers against shared perturbations. Moreover, it is particularly effective in removing universal perturbations, which can be seen as an extreme form of shared perturbations. Unfortunately, adversarial training does not consistently increase the robustness against singular perturbations on unseen inputs. However, we find that adversarial training decreases robustness of the remaining perturbations against image transformations such as changes to contrast and brightness or Gaussian blurring. It thus makes successful attacks on the classifier in the physical world less likely. Finally, we show that even singular perturbations can be easily detected and must thus exhibit generalizable patterns even though the perturbations are specific for certain inputs. |
| Tasks | |
| Published | 2018-01-01 |
| URL | https://openreview.net/forum?id=SyjsLqxR- |
| https://openreview.net/pdf?id=SyjsLqxR- | |
| PWC | https://paperswithcode.com/paper/universality-robustness-and-detectability-of |
| Repo | |
| Framework | |
Development of a Mobile Observation Support System for Students: FishWatchr Mini
| Title | Development of a Mobile Observation Support System for Students: FishWatchr Mini |
| Authors | Masaya Yamaguchi, Masanori Kitamura, Naomi Yanagida |
| Abstract | |
| Tasks | |
| Published | 2018-05-01 |
| URL | https://www.aclweb.org/anthology/L18-1359/ |
| https://www.aclweb.org/anthology/L18-1359 | |
| PWC | https://paperswithcode.com/paper/development-of-a-mobile-observation-support |
| Repo | |
| Framework | |
An Evaluation of Image-Based Verb Prediction Models against Human Eye-Tracking Data
| Title | An Evaluation of Image-Based Verb Prediction Models against Human Eye-Tracking Data |
| Authors | Sp Gella, ana, Frank Keller |
| Abstract | Recent research in language and vision has developed models for predicting and disambiguating verbs from images. Here, we ask whether the predictions made by such models correspond to human intuitions about visual verbs. We show that the image regions a verb prediction model identifies as salient for a given verb correlate with the regions fixated by human observers performing a verb classification task. |
| Tasks | Eye Tracking, Question Answering, Visual Question Answering, Word Sense Disambiguation |
| Published | 2018-06-01 |
| URL | https://www.aclweb.org/anthology/N18-2119/ |
| https://www.aclweb.org/anthology/N18-2119 | |
| PWC | https://paperswithcode.com/paper/an-evaluation-of-image-based-verb-prediction |
| Repo | |
| Framework | |
Event Time Extraction with a Decision Tree of Neural Classifiers
| Title | Event Time Extraction with a Decision Tree of Neural Classifiers |
| Authors | Nils Reimers, Nazanin Dehghani, Iryna Gurevych |
| Abstract | Extracting the information from text when an event happened is challenging. Documents do not only report on current events, but also on past events as well as on future events. Often, the relevant time information for an event is scattered across the document. In this paper we present a novel method to automatically anchor events in time. To our knowledge it is the first approach that takes temporal information from the complete document into account. We created a decision tree that applies neural network based classifiers at its nodes. We use this tree to incrementally infer, in a stepwise manner, at which time frame an event happened. We evaluate the approach on the TimeBank-EventTime Corpus (Reimers et al., 2016) achieving an accuracy of 42.0{%} compared to an inter-annotator agreement (IAA) of 56.7{%}. For events that span over a single day we observe an accuracy improvement of 33.1 points compared to the state-of-the-art CAEVO system (Chambers et al., 2014). Without retraining, we apply this model to the SemEval-2015 Task 4 on automatic timeline generation and achieve an improvement of 4.01 points F1-score compared to the state-of-the-art. Our code is publically available. |
| Tasks | |
| Published | 2018-01-01 |
| URL | https://www.aclweb.org/anthology/Q18-1006/ |
| https://www.aclweb.org/anthology/Q18-1006 | |
| PWC | https://paperswithcode.com/paper/event-time-extraction-with-a-decision-tree-of |
| Repo | |
| Framework | |
The Emergence of Semantics in Neural Network Representations of Visual Information
| Title | The Emergence of Semantics in Neural Network Representations of Visual Information |
| Authors | Dhanush Dharmaretnam, Alona Fyshe |
| Abstract | Word vector models learn about semantics through corpora. Convolutional Neural Networks (CNNs) can learn about semantics through images. At the most abstract level, some of the information in these models must be shared, as they model the same real-world phenomena. Here we employ techniques previously used to detect semantic representations in the human brain to detect semantic representations in CNNs. We show the accumulation of semantic information in the layers of the CNN, and discover that, for misclassified images, the correct class can be recovered in intermediate layers of a CNN. |
| Tasks | |
| Published | 2018-06-01 |
| URL | https://www.aclweb.org/anthology/N18-2122/ |
| https://www.aclweb.org/anthology/N18-2122 | |
| PWC | https://paperswithcode.com/paper/the-emergence-of-semantics-in-neural-network |
| Repo | |
| Framework | |
Signbank: Software to Support Web Based Dictionaries of Sign Language
| Title | Signbank: Software to Support Web Based Dictionaries of Sign Language |
| Authors | Steve Cassidy, Onno Crasborn, Henri Nieminen, Wessel Stoop, Micha Hulsbosch, Susan Even, Erwin Komen, Trevor Johnston |
| Abstract | |
| Tasks | |
| Published | 2018-05-01 |
| URL | https://www.aclweb.org/anthology/L18-1374/ |
| https://www.aclweb.org/anthology/L18-1374 | |
| PWC | https://paperswithcode.com/paper/signbank-software-to-support-web-based |
| Repo | |
| Framework | |
Distribution Regression Network
| Title | Distribution Regression Network |
| Authors | Connie Kou, Hwee Kuan Lee, Teck Khim Ng |
| Abstract | We introduce our Distribution Regression Network (DRN) which performs regression from input probability distributions to output probability distributions. Compared to existing methods, DRN learns with fewer model parameters and easily extends to multiple input and multiple output distributions. On synthetic and real-world datasets, DRN performs similarly or better than the state-of-the-art. Furthermore, DRN generalizes the conventional multilayer perceptron (MLP). In the framework of MLP, each node encodes a real number, whereas in DRN, each node encodes a probability distribution. |
| Tasks | |
| Published | 2018-01-01 |
| URL | https://openreview.net/forum?id=ByYPLJA6W |
| https://openreview.net/pdf?id=ByYPLJA6W | |
| PWC | https://paperswithcode.com/paper/distribution-regression-network |
| Repo | |
| Framework | |
GLoMo: Unsupervised Learning of Transferable Relational Graphs
| Title | GLoMo: Unsupervised Learning of Transferable Relational Graphs |
| Authors | Zhilin Yang, Jake Zhao, Bhuwan Dhingra, Kaiming He, William W. Cohen, Ruslan R. Salakhutdinov, Yann Lecun |
| Abstract | Modern deep transfer learning approaches have mainly focused on learning generic feature vectors from one task that are transferable to other tasks, such as word embeddings in language and pretrained convolutional features in vision. However, these approaches usually transfer unary features and largely ignore more structured graphical representations. This work explores the possibility of learning generic latent relational graphs that capture dependencies between pairs of data units (e.g., words or pixels) from large-scale unlabeled data and transferring the graphs to downstream tasks. Our proposed transfer learning framework improves performance on various tasks including question answering, natural language inference, sentiment analysis, and image classification. We also show that the learned graphs are generic enough to be transferred to different embeddings on which the graphs have not been trained (including GloVe embeddings, ELMo embeddings, and task-specific RNN hidden units), or embedding-free units such as image pixels. |
| Tasks | Image Classification, Natural Language Inference, Question Answering, Sentiment Analysis, Transfer Learning, Word Embeddings |
| Published | 2018-12-01 |
| URL | http://papers.nips.cc/paper/8110-glomo-unsupervised-learning-of-transferable-relational-graphs |
| http://papers.nips.cc/paper/8110-glomo-unsupervised-learning-of-transferable-relational-graphs.pdf | |
| PWC | https://paperswithcode.com/paper/glomo-unsupervised-learning-of-transferable |
| Repo | |
| Framework | |
EMA at SemEval-2018 Task 1: Emotion Mining for Arabic
| Title | EMA at SemEval-2018 Task 1: Emotion Mining for Arabic |
| Authors | Gilbert Badaro, Obeida El Jundi, Alaa Khaddaj, Alaa Maarouf, Raslan Kain, Hazem Hajj, Wassim El-Hajj |
| Abstract | While significant progress has been achieved for Opinion Mining in Arabic (OMA), very limited efforts have been put towards the task of Emotion mining in Arabic. In fact, businesses are interested in learning a fine-grained representation of how users are feeling towards their products or services. In this work, we describe the methods used by the team Emotion Mining in Arabic (EMA), as part of the SemEval-2018 Task 1 for Affect Mining for Arabic tweets. EMA participated in all 5 subtasks. For the five tasks, several preprocessing steps were evaluated and eventually the best system included diacritics removal, elongation adjustment, replacement of emojis by the corresponding Arabic word, character normalization and light stemming. Moreover, several features were evaluated along with different classification and regression techniques. For the 5 subtasks, word embeddings feature turned out to perform best along with Ensemble technique. EMA achieved the 1st place in subtask 5, and 3rd place in subtasks 1 and 3. |
| Tasks | Emotion Classification, Emotion Recognition, Opinion Mining, Recommendation Systems, Sentiment Analysis, Word Embeddings |
| Published | 2018-06-01 |
| URL | https://www.aclweb.org/anthology/S18-1036/ |
| https://www.aclweb.org/anthology/S18-1036 | |
| PWC | https://paperswithcode.com/paper/ema-at-semeval-2018-task-1-emotion-mining-for |
| Repo | |
| Framework | |
Grassmann Pooling as Compact Homogeneous Bilinear Pooling for Fine-Grained Visual Classification
| Title | Grassmann Pooling as Compact Homogeneous Bilinear Pooling for Fine-Grained Visual Classification |
| Authors | Xing Wei, Yue Zhang, Yihong Gong, Jiawei Zhang, Nanning Zheng |
| Abstract | Designing discriminative and invariant features is the key to visual recognition. Recently, the bilinear pooled feature matrix of Convolutional Neural Network (CNN) has shown to achieve state-of-the-art performance on a range of fine-grained visual recognition tasks. The bilinear feature matrix collects second-order statistics and is closely related to the covariance matrix descriptor. However, the bilinear feature could suffer from the visual burstiness phenomenon similar to other visual representations such as VLAD and Fisher Vector. The reason is that the bilinear feature matrix is sensitive to the magnitudes and correlations of local CNN feature elements which can be measured by its singular values. On the other hand, the singular vectors are more invariant and reasonable to be adopted as the feature representation. Motivated by this point, we advocate an alternative pooling method which transforms the CNN feature matrix to an orthonormal matrix consists of its principal singular vectors. Geometrically, such orthonormal matrix lies on the Grassmann manifold, a Riemannian manifold whose points represent subspaces of the Euclidean space. Similarity measurement of images reduces to comparing the principal angles between these ``homogeneous” subspaces and thus is independent of the magnitudes and correlations of local CNN activations. In particular, we demonstrate that the projection distance on the Grassmann manifold deduces a bilinear feature mapping without explicitly computing the bilinear feature matrix, which enables a very compact feature and classifier representation. Experimental results show that our method achieves an excellent balance of model complexity and accuracy on a variety of fine-grained image classification datasets. | |
| Tasks | Fine-Grained Image Classification, Fine-Grained Visual Recognition, Image Classification |
| Published | 2018-09-01 |
| URL | http://openaccess.thecvf.com/content_ECCV_2018/html/Xing_Wei_Grassmann_Pooling_for_ECCV_2018_paper.html |
| http://openaccess.thecvf.com/content_ECCV_2018/papers/Xing_Wei_Grassmann_Pooling_for_ECCV_2018_paper.pdf | |
| PWC | https://paperswithcode.com/paper/grassmann-pooling-as-compact-homogeneous |
| Repo | |
| Framework | |
J-MeDic: A Japanese Disease Name Dictionary based on Real Clinical Usage
| Title | J-MeDic: A Japanese Disease Name Dictionary based on Real Clinical Usage |
| Authors | Kaoru Ito, Hiroyuki Nagai, Taro Okahisa, Shoko Wakamiya, Tomohide Iwao, Eiji Aramaki |
| Abstract | |
| Tasks | Named Entity Recognition |
| Published | 2018-05-01 |
| URL | https://www.aclweb.org/anthology/L18-1375/ |
| https://www.aclweb.org/anthology/L18-1375 | |
| PWC | https://paperswithcode.com/paper/j-medic-a-japanese-disease-name-dictionary |
| Repo | |
| Framework | |
Corpus-based Content Construction
| Title | Corpus-based Content Construction |
| Authors | Balaji Vasan Srinivasan, Pranav Maneriker, Kundan Krishna, Natwar Modani |
| Abstract | Enterprise content writers are engaged in writing textual content for various purposes. Often, the text being written may already be present in the enterprise corpus in the form of past articles and can be re-purposed for the current needs. In the absence of suitable tools, authors manually curate/create such content (sometimes from scratch) which reduces their productivity. To address this, we propose an automatic approach to generate an initial version of the author{'}s intended text based on an input content snippet. Starting with a set of extracted textual fragments related to the snippet based on the query words in it, the proposed approach builds the desired text from these fragment by simultaneously optimizing the information coverage, relevance, diversity and coherence in the generated content. Evaluations on standard datasets shows improved performance against existing baselines on several metrics. |
| Tasks | |
| Published | 2018-08-01 |
| URL | https://www.aclweb.org/anthology/C18-1297/ |
| https://www.aclweb.org/anthology/C18-1297 | |
| PWC | https://paperswithcode.com/paper/corpus-based-content-construction |
| Repo | |
| Framework | |
