Paper Group NANR 238
Label-PEnet: Sequential Label Propagation and Enhancement Networks for Weakly Supervised Instance Segmentation. On Efficient Retrieval of Top Similarity Vectors. Polar Prototype Networks. Evaluating Automatic Term Extraction Methods on Individual Documents. Numbers Normalisation in the Inflected Languages: a Case Study of Polish. Riemannian TransE: …
Label-PEnet: Sequential Label Propagation and Enhancement Networks for Weakly Supervised Instance Segmentation
| Title | Label-PEnet: Sequential Label Propagation and Enhancement Networks for Weakly Supervised Instance Segmentation |
| Authors | Weifeng Ge, Sheng Guo, Weilin Huang, Matthew R. Scott |
| Abstract | Weakly-supervised instance segmentation aims to detect and segment object instances precisely, given image-level labels only. Unlike previous methods which are composed of multiple offline stages, we propose Sequential Label Propagation and Enhancement Networks (referred as Label-PEnet) that progressively transforms image-level labels to pixel-wise labels in a coarse-to-fine manner. We design four cascaded modules including multi-label classification, object detection, instance refinement and instance segmentation, which are implemented sequentially by sharing the same backbone. The cascaded pipeline is trained alternatively with a curriculum learning strategy that generalizes labels from high level images to low-level pixels gradually with increasing accuracy. In addition, we design a proposal calibration module to explore the ability of classification networks to find key pixels that identify object parts, which serves as a post validation strategy running in the inverse order. We evaluate the efficiency of our Label-PEnet in mining instance masks on standard benchmarks: PASCAL VOC 2007 and 2012. Experimental results show that Label-PEnet outperforms the state-of-art algorithms by a clear margin, and obtains comparable performance even with fully supervised approaches. |
| Tasks | Calibration, Instance Segmentation, Multi-Label Classification, Object Detection, Semantic Segmentation, Weakly-supervised instance segmentation |
| Published | 2019-10-01 |
| URL | http://openaccess.thecvf.com/content_ICCV_2019/html/Ge_Label-PEnet_Sequential_Label_Propagation_and_Enhancement_Networks_for_Weakly_Supervised_ICCV_2019_paper.html |
| http://openaccess.thecvf.com/content_ICCV_2019/papers/Ge_Label-PEnet_Sequential_Label_Propagation_and_Enhancement_Networks_for_Weakly_Supervised_ICCV_2019_paper.pdf | |
| PWC | https://paperswithcode.com/paper/label-penet-sequential-label-propagation-and-1 |
| Repo | |
| Framework | |
On Efficient Retrieval of Top Similarity Vectors
| Title | On Efficient Retrieval of Top Similarity Vectors |
| Authors | Shulong Tan, Zhixin Zhou, Zhaozhuo Xu, Ping Li |
| Abstract | Retrieval of relevant vectors produced by representation learning critically influences the efficiency in natural language processing (NLP) tasks. In this paper, we demonstrate an efficient method for searching vectors via a typical non-metric matching function: inner product. Our method, which constructs an approximate Inner Product Delaunay Graph (IPDG) for top-1 Maximum Inner Product Search (MIPS), transforms retrieving the most suitable latent vectors into a graph search problem with great benefits of efficiency. Experiments on data representations learned for different machine learning tasks verify the outperforming effectiveness and efficiency of the proposed IPDG. |
| Tasks | Representation Learning |
| Published | 2019-11-01 |
| URL | https://www.aclweb.org/anthology/D19-1527/ |
| https://www.aclweb.org/anthology/D19-1527 | |
| PWC | https://paperswithcode.com/paper/on-efficient-retrieval-of-top-similarity |
| Repo | |
| Framework | |
Polar Prototype Networks
| Title | Polar Prototype Networks |
| Authors | Pascal Mettes, Elise van der Pol, Cees G. M. Snoek |
| Abstract | This paper proposes a neural network for classification and regression, without the need to learn layout structures in the output space. Standard solutions such as softmax cross-entropy and mean squared error are effective but parametric, meaning that known inductive structures such as maximum margin separation and simplicity (Occam’s Razor) need to be learned for the task at hand. Instead, we propose polar prototype networks, a class of networks that explicitly states the structure, \ie the layout, of the output. The structure is defined by polar prototypes, points on the hypersphere of the output space. For classification, each class is described by a single polar prototype and they are a priori distributed with maximal separation and equal shares on the hypersphere. Classes are assigned to prototypes randomly or based on semantic priors and training becomes a matter of minimizing angular distances between examples and their class prototypes. For regression, we show that training can be performed as a polar interpolation between two prototypes, arriving at a regression with higher-dimensional outputs. From empirical analysis, we find that polar prototype networks benefit from large margin separation and semantic class structure, while only requiring a minimal amount of output dimensions. While the structure is simple, the performance is on par with (classification) or better than (regression) standard network methods. Moreover, we show that we gain the ability to perform regression and classification jointly in the same space, which is disentangled and interpretable by design. |
| Tasks | |
| Published | 2019-05-01 |
| URL | https://openreview.net/forum?id=Syx4_iCqKQ |
| https://openreview.net/pdf?id=Syx4_iCqKQ | |
| PWC | https://paperswithcode.com/paper/polar-prototype-networks |
| Repo | |
| Framework | |
Evaluating Automatic Term Extraction Methods on Individual Documents
| Title | Evaluating Automatic Term Extraction Methods on Individual Documents |
| Authors | Antonio {\v{S}}ajatovi{'c}, Maja Buljan, Jan {\v{S}}najder, Bojana Dalbelo Ba{\v{s}}i{'c} |
| Abstract | Automatic Term Extraction (ATE) extracts terminology from domain-specific corpora. ATE is used in many NLP tasks, including Computer Assisted Translation, where it is typically applied to individual documents rather than the entire corpus. While corpus-level ATE has been extensively evaluated, it is not obvious how the results transfer to document-level ATE. To fill this gap, we evaluate 16 state-of-the-art ATE methods on full-length documents from three different domains, on both corpus and document levels. Unlike existing studies, our evaluation is more realistic as we take into account all gold terms. We show that no single method is best in corpus-level ATE, but C-Value and KeyConceptRelatendess surpass others in document-level ATE. |
| Tasks | |
| Published | 2019-08-01 |
| URL | https://www.aclweb.org/anthology/W19-5118/ |
| https://www.aclweb.org/anthology/W19-5118 | |
| PWC | https://paperswithcode.com/paper/evaluating-automatic-term-extraction-methods |
| Repo | |
| Framework | |
Numbers Normalisation in the Inflected Languages: a Case Study of Polish
| Title | Numbers Normalisation in the Inflected Languages: a Case Study of Polish |
| Authors | Rafa{\l} Po{'s}wiata, Micha{\l} Pere{\l}kiewicz |
| Abstract | Text normalisation in Text-to-Speech systems is a process of converting written expressions to their spoken forms. This task is complicated because in many cases the normalised form depends on the context. Furthermore, when we analysed languages like Croatian, Lithuanian, Polish, Russian or Slovak there is additional difficulty related to their inflected nature. In this paper we want to show how to deal with this problem for one of these languages: Polish, without having a large dedicated data set and using solutions prepared for other NLP tasks. We limited our study to only numbers expressions, which are the most common non-standard words to normalise. The proposed solution is a combination of morphological tagger and transducer supported by a dictionary of numbers in their spoken forms. The data set used for evaluation is based on the part of 1-million word subset of the National Corpus of Polish. The accuracy of the described approach is presented with a comparison to a simple baseline and two commercial systems: Google Cloud Text-to-Speech and Amazon Polly. |
| Tasks | |
| Published | 2019-08-01 |
| URL | https://www.aclweb.org/anthology/W19-3703/ |
| https://www.aclweb.org/anthology/W19-3703 | |
| PWC | https://paperswithcode.com/paper/numbers-normalisation-in-the-inflected |
| Repo | |
| Framework | |
Riemannian TransE: Multi-relational Graph Embedding in Non-Euclidean Space
| Title | Riemannian TransE: Multi-relational Graph Embedding in Non-Euclidean Space |
| Authors | Atsushi Suzuki, Yosuke Enokida, Kenji Yamanishi |
| Abstract | Multi-relational graph embedding which aims at achieving effective representations with reduced low-dimensional parameters, has been widely used in knowledge base completion. Although knowledge base data usually contains tree-like or cyclic structure, none of existing approaches can embed these data into a compatible space that in line with the structure. To overcome this problem, a novel framework, called Riemannian TransE, is proposed in this paper to embed the entities in a Riemannian manifold. Riemannian TransE models each relation as a move to a point and defines specific novel distance dissimilarity for each relation, so that all the relations are naturally embedded in correspondence to the structure of data. Experiments on several knowledge base completion tasks have shown that, based on an appropriate choice of manifold, Riemannian TransE achieves good performance even with a significantly reduced parameters. |
| Tasks | Graph Embedding, Knowledge Base Completion |
| Published | 2019-05-01 |
| URL | https://openreview.net/forum?id=r1xRW3A9YX |
| https://openreview.net/pdf?id=r1xRW3A9YX | |
| PWC | https://paperswithcode.com/paper/riemannian-transe-multi-relational-graph |
| Repo | |
| Framework | |
Weakly Supervised Attentional Model for Low Resource Ad-hoc Cross-lingual Information Retrieval
| Title | Weakly Supervised Attentional Model for Low Resource Ad-hoc Cross-lingual Information Retrieval |
| Authors | Lingjun Zhao, Rabih Zbib, Zhuolin Jiang, Damianos Karakos, Zhongqiang Huang |
| Abstract | We propose a weakly supervised neural model for Ad-hoc Cross-lingual Information Retrieval (CLIR) from low-resource languages. Low resource languages often lack relevance annotations for CLIR, and when available the training data usually has limited coverage for possible queries. In this paper, we design a model which does not require relevance annotations, instead it is trained on samples extracted from translation corpora as weak supervision. This model relies on an attention mechanism to learn spans in the foreign sentence that are relevant to the query. We report experiments on two low resource languages: Swahili and Tagalog, trained on less that 100k parallel sentences each. The proposed model achieves 19 MAP points improvement compared to using CNNs for feature extraction, 12 points improvement from machine translation-based CLIR, and up to 6 points improvement compared to probabilistic CLIR models. |
| Tasks | Information Retrieval, Machine Translation |
| Published | 2019-11-01 |
| URL | https://www.aclweb.org/anthology/D19-6129/ |
| https://www.aclweb.org/anthology/D19-6129 | |
| PWC | https://paperswithcode.com/paper/weakly-supervised-attentional-model-for-low |
| Repo | |
| Framework | |
At a Glance: The Impact of Gaze Aggregation Views on Syntactic Tagging
| Title | At a Glance: The Impact of Gaze Aggregation Views on Syntactic Tagging |
| Authors | Sigrid Klerke, Barbara Plank |
| Abstract | Readers{'} eye movements used as part of the training signal have been shown to improve performance in a wide range of Natural Language Processing (NLP) tasks. Previous work uses gaze data either at the type level or at the token level and mostly from a single eye-tracking corpus. In this paper, we analyze type vs token-level integration options with eye tracking data from two corpora to inform two syntactic sequence labeling problems: binary phrase chunking and part-of-speech tagging. We show that using globally-aggregated measures that capture the central tendency or variability of gaze data is more beneficial than proposed local views which retain individual participant information. While gaze data is informative for supervised POS tagging, which complements previous findings on unsupervised POS induction, almost no improvement is obtained for binary phrase chunking, except for a single specific setup. Hence, caution is warranted when using gaze data as signal for NLP, as no single view is robust over tasks, modeling choice and gaze corpus. |
| Tasks | Chunking, Eye Tracking, Part-Of-Speech Tagging |
| Published | 2019-11-01 |
| URL | https://www.aclweb.org/anthology/D19-6408/ |
| https://www.aclweb.org/anthology/D19-6408 | |
| PWC | https://paperswithcode.com/paper/at-a-glance-the-impact-of-gaze-aggregation |
| Repo | |
| Framework | |
Assessing Personally Perceived Image Quality via Image Features and Collaborative Filtering
| Title | Assessing Personally Perceived Image Quality via Image Features and Collaborative Filtering |
| Authors | Jari Korhonen |
| Abstract | During the past few years, different methods for optimizing the camera settings and post-processing techniques to improve the subjective quality of consumer photos have been studied extensively. However, most of the research in the prior art has focused on finding the optimal method for an average user. Since there is large deviation in personal opinions and aesthetic standards, the next challenge is to find the settings and post-processing techniques that fit to the individual users’ personal taste. In this study, we aim to predict the personally perceived image quality by combining classical image feature analysis and collaboration filtering approach known from the recommendation systems. The experimental results for the proposed method show promising results. As a practical application, our work can be used for personalizing the camera settings or post-processing parameters for different users and images. |
| Tasks | Recommendation Systems |
| Published | 2019-06-01 |
| URL | http://openaccess.thecvf.com/content_CVPR_2019/html/Korhonen_Assessing_Personally_Perceived_Image_Quality_via_Image_Features_and_Collaborative_CVPR_2019_paper.html |
| http://openaccess.thecvf.com/content_CVPR_2019/papers/Korhonen_Assessing_Personally_Perceived_Image_Quality_via_Image_Features_and_Collaborative_CVPR_2019_paper.pdf | |
| PWC | https://paperswithcode.com/paper/assessing-personally-perceived-image-quality |
| Repo | |
| Framework | |
Guaranteed Matrix Completion Under Multiple Linear Transformations
| Title | Guaranteed Matrix Completion Under Multiple Linear Transformations |
| Authors | Chao Li, Wei He, Longhao Yuan, Zhun Sun, Qibin Zhao |
| Abstract | Low-rank matrix completion (LRMC) is a classical model in both computer vision (CV) and machine learning, and has been successfully applied to various real applications. In the recent CV tasks, the completion is usually employed on the variants of data, such as “non-local” or filtered, rather than their original forms. This fact makes that the theoretical analysis of the conventional LRMC is no longer suitable in these applications. To tackle this problem, we propose a more general framework for LRMC, in which the linear transformations of the data are taken into account. We rigorously prove the identifiability of the proposed model and show an upper bound of the reconstruction error. Furthermore, we derive an efficient completion algorithm by using augmented Lagrangian multipliers and the sketching trick. In the experiments, we apply the proposed method to the classical image inpainting problem and achieve the state-of-the-art results. |
| Tasks | Image Inpainting, Low-Rank Matrix Completion, Matrix Completion |
| Published | 2019-06-01 |
| URL | http://openaccess.thecvf.com/content_CVPR_2019/html/Li_Guaranteed_Matrix_Completion_Under_Multiple_Linear_Transformations_CVPR_2019_paper.html |
| http://openaccess.thecvf.com/content_CVPR_2019/papers/Li_Guaranteed_Matrix_Completion_Under_Multiple_Linear_Transformations_CVPR_2019_paper.pdf | |
| PWC | https://paperswithcode.com/paper/guaranteed-matrix-completion-under-multiple |
| Repo | |
| Framework | |
PEPSI : Fast Image Inpainting With Parallel Decoding Network
| Title | PEPSI : Fast Image Inpainting With Parallel Decoding Network |
| Authors | Min-cheol Sagong, Yong-goo Shin, Seung-wook Kim, Seung Park, Sung-jea Ko |
| Abstract | Recently, a generative adversarial network (GAN)-based method employing the coarse-to-fine network with the contextual attention module (CAM) has shown outstanding results in image inpainting. However, this method requires numerous computational resources due to its two-stage process for feature encoding. To solve this problem, in this paper, we present a novel network structure, called PEPSI: parallel extended-decoder path for semantic inpainting. PEPSI can reduce the number of convolution operations by adopting a structure consisting of a single shared encoding network and a parallel decoding network with coarse and inpainting paths. The coarse path produces a preliminary inpainting result with which the encoding network is trained to predict features for the CAM. At the same time, the inpainting path creates a higher-quality inpainting result using refined features reconstructed by the CAM. PEPSI not only reduces the number of convolution operation almost by half as compared to the conventional coarse-to-fine networks but also exhibits superior performance to other models in terms of testing time and qualitative scores. |
| Tasks | Image Inpainting |
| Published | 2019-06-01 |
| URL | http://openaccess.thecvf.com/content_CVPR_2019/html/Sagong_PEPSI__Fast_Image_Inpainting_With_Parallel_Decoding_Network_CVPR_2019_paper.html |
| http://openaccess.thecvf.com/content_CVPR_2019/papers/Sagong_PEPSI__Fast_Image_Inpainting_With_Parallel_Decoding_Network_CVPR_2019_paper.pdf | |
| PWC | https://paperswithcode.com/paper/pepsi-fast-image-inpainting-with-parallel |
| Repo | |
| Framework | |
Sakura: Large-scale Incorrect Example Retrieval System for Learners of Japanese as a Second Language
| Title | Sakura: Large-scale Incorrect Example Retrieval System for Learners of Japanese as a Second Language |
| Authors | Mio Arai, Tomonori Kodaira, Mamoru Komachi |
| Abstract | This study develops an incorrect example retrieval system, called Sakura, using a large-scale Lang-8 dataset for Japanese language learners. Existing example retrieval systems do not include grammatically incorrect examples or present only a few examples, if any. If a retrieval system has a wide coverage of incorrect examples along with the correct counterpart, learners can revise their composition themselves. Considering the usability of retrieving incorrect examples, our proposed system uses a large-scale corpus to expand the coverage of incorrect examples and presents correct expressions along with incorrect expressions. Our intrinsic and extrinsic evaluations indicate that our system is more useful than a previous system. |
| Tasks | |
| Published | 2019-07-01 |
| URL | https://www.aclweb.org/anthology/P19-3001/ |
| https://www.aclweb.org/anthology/P19-3001 | |
| PWC | https://paperswithcode.com/paper/sakura-large-scale-incorrect-example |
| Repo | |
| Framework | |
Deep Restoration of Vintage Photographs From Scanned Halftone Prints
| Title | Deep Restoration of Vintage Photographs From Scanned Halftone Prints |
| Authors | Qifan Gao, Xiao Shu, Xiaolin Wu |
| Abstract | A great number of invaluable historical photographs unfortunately only exist in the form of halftone prints in old publications such as newspapers or books. Their original continuous-tone films have long been lost or irreparably damaged. There have been attempts to digitally restore these vintage halftone prints to the original film quality or higher. However, even using powerful deep convolutional neural networks, it is still difficult to obtain satisfactory results. The main challenge is that the degradation process is complex and compounded while little to no real data is available for properly training a data-driven method. In this research, we adopt a novel strategy of two-stage deep learning, in which the restoration task is divided into two stages: the removal of printing artifacts and the inverse of halftoning. The advantage of our technique is that only the simple first stage requires unsupervised training in order to make the combined network generalize on real halftone prints, while the more complex second stage of inverse halftoning can be easily trained with synthetic data. Extensive experimental results demonstrate the efficacy of the proposed technique for real halftone prints; the new technique significantly outperforms the existing ones in visual quality. |
| Tasks | |
| Published | 2019-10-01 |
| URL | http://openaccess.thecvf.com/content_ICCV_2019/html/Gao_Deep_Restoration_of_Vintage_Photographs_From_Scanned_Halftone_Prints_ICCV_2019_paper.html |
| http://openaccess.thecvf.com/content_ICCV_2019/papers/Gao_Deep_Restoration_of_Vintage_Photographs_From_Scanned_Halftone_Prints_ICCV_2019_paper.pdf | |
| PWC | https://paperswithcode.com/paper/deep-restoration-of-vintage-photographs-from |
| Repo | |
| Framework | |
Modeling Document-level Causal Structures for Event Causal Relation Identification
| Title | Modeling Document-level Causal Structures for Event Causal Relation Identification |
| Authors | Lei Gao, Prafulla Kumar Choubey, Ruihong Huang |
| Abstract | We aim to comprehensively identify all the event causal relations in a document, both within a sentence and across sentences, which is important for reconstructing pivotal event structures. The challenges we identified are two: 1) event causal relations are sparse among all possible event pairs in a document, in addition, 2) few causal relations are explicitly stated. Both challenges are especially true for identifying causal relations between events across sentences. To address these challenges, we model rich aspects of document-level causal structures for achieving comprehensive causal relation identification. The causal structures include heavy involvements of document-level main events in causal relations as well as several types of fine-grained constraints that capture implications from certain sentential syntactic relations and discourse relations as well as interactions between event causal relations and event coreference relations. Our experimental results show that modeling the global and fine-grained aspects of causal structures using Integer Linear Programming (ILP) greatly improves the performance of causal relation identification, especially in identifying cross-sentence causal relations. |
| Tasks | |
| Published | 2019-06-01 |
| URL | https://www.aclweb.org/anthology/N19-1179/ |
| https://www.aclweb.org/anthology/N19-1179 | |
| PWC | https://paperswithcode.com/paper/modeling-document-level-causal-structures-for |
| Repo | |
| Framework | |
Enhancing Local Feature Extraction with Global Representation for Neural Text Classification
| Title | Enhancing Local Feature Extraction with Global Representation for Neural Text Classification |
| Authors | Guocheng Niu, Hengru Xu, Bolei He, Xinyan Xiao, Hua Wu, Sheng Gao |
| Abstract | For text classification, traditional local feature driven models learn long dependency by deeply stacking or hybrid modeling. This paper proposes a novel Encoder1-Encoder2 architecture, where global information is incorporated into the procedure of local feature extraction from scratch. In particular, Encoder1 serves as a global information provider, while Encoder2 performs as a local feature extractor and is directly fed into the classifier. Meanwhile, two modes are also designed for their interaction. Thanks to the awareness of global information, our method is able to learn better instance specific local features and thus avoids complicated upper operations. Experiments conducted on eight benchmark datasets demonstrate that our proposed architecture promotes local feature driven models by a substantial margin and outperforms the previous best models in the fully-supervised setting. |
| Tasks | Text Classification |
| Published | 2019-11-01 |
| URL | https://www.aclweb.org/anthology/D19-1047/ |
| https://www.aclweb.org/anthology/D19-1047 | |
| PWC | https://paperswithcode.com/paper/enhancing-local-feature-extraction-with |
| Repo | |
| Framework | |
