Paper Group NANR 6
An End-to-End Chinese Discourse Parser with Adaptation to Explicit and Non-explicit Relation Recognition. Short-Dot: Computing Large Linear Transforms Distributedly Using Coded Short Dot Products. A primal-dual method for conic constrained distributed optimization problems. MUTT: Metric Unit TesTing for Language Generation Tasks. Dual Decomposed Le …
An End-to-End Chinese Discourse Parser with Adaptation to Explicit and Non-explicit Relation Recognition
| Title | An End-to-End Chinese Discourse Parser with Adaptation to Explicit and Non-explicit Relation Recognition |
| Authors | Xiaomian Kang, Haoran Li, Long Zhou, Jiajun Zhang, Chengqing Zong |
| Abstract | |
| Tasks | Machine Translation, Question Answering, Text Summarization |
| Published | 2016-08-01 |
| URL | https://www.aclweb.org/anthology/K16-2003/ |
| https://www.aclweb.org/anthology/K16-2003 | |
| PWC | https://paperswithcode.com/paper/an-end-to-end-chinese-discourse-parser-with |
| Repo | |
| Framework | |
Short-Dot: Computing Large Linear Transforms Distributedly Using Coded Short Dot Products
| Title | Short-Dot: Computing Large Linear Transforms Distributedly Using Coded Short Dot Products |
| Authors | Sanghamitra Dutta, Viveck Cadambe, Pulkit Grover |
| Abstract | Faced with saturation of Moore’s law and increasing size and dimension of data, system designers have increasingly resorted to parallel and distributed computing to reduce computation time of machine-learning algorithms. However, distributed computing is often bottle necked by a small fraction of slow processors called “stragglers” that reduce the speed of computation because the fusion node has to wait for all processors to complete their processing. To combat the effect of stragglers, recent literature proposes introducing redundancy in computations across processors, e.g., using repetition-based strategies or erasure codes. The fusion node can exploit this redundancy by completing the computation using outputs from only a subset of the processors, ignoring the stragglers. In this paper, we propose a novel technique - that we call “Short-Dot” - to introduce redundant computations in a coding theory inspired fashion, for computing linear transforms of long vectors. Instead of computing long dot products as required in the original linear transform, we construct a larger number of redundant and short dot products that can be computed more efficiently at individual processors. Further, only a subset of these short dot products are required at the fusion node to finish the computation successfully. We demonstrate through probabilistic analysis as well as experiments on computing clusters that Short-Dot offers significant speed-up compared to existing techniques. We also derive trade-offs between the length of the dot-products and the resilience to stragglers (number of processors required to finish), for any such strategy and compare it to that achieved by our strategy. |
| Tasks | |
| Published | 2016-12-01 |
| URL | http://papers.nips.cc/paper/6329-short-dot-computing-large-linear-transforms-distributedly-using-coded-short-dot-products |
| http://papers.nips.cc/paper/6329-short-dot-computing-large-linear-transforms-distributedly-using-coded-short-dot-products.pdf | |
| PWC | https://paperswithcode.com/paper/short-dot-computing-large-linear-transforms |
| Repo | |
| Framework | |
A primal-dual method for conic constrained distributed optimization problems
| Title | A primal-dual method for conic constrained distributed optimization problems |
| Authors | Necdet Serhat Aybat, Erfan Yazdandoost Hamedani |
| Abstract | We consider cooperative multi-agent consensus optimization problems over an undirected network of agents, where only those agents connected by an edge can directly communicate. The objective is to minimize the sum of agent-specific composite convex functions over agent-specific private conic constraint sets; hence, the optimal consensus decision should lie in the intersection of these private sets. We provide convergence rates in sub-optimality, infeasibility and consensus violation; examine the effect of underlying network topology on the convergence rates of the proposed decentralized algorithms; and show how to extend these methods to handle time-varying communication networks. |
| Tasks | Distributed Optimization |
| Published | 2016-12-01 |
| URL | http://papers.nips.cc/paper/6242-a-primal-dual-method-for-conic-constrained-distributed-optimization-problems |
| http://papers.nips.cc/paper/6242-a-primal-dual-method-for-conic-constrained-distributed-optimization-problems.pdf | |
| PWC | https://paperswithcode.com/paper/a-primal-dual-method-for-conic-constrained |
| Repo | |
| Framework | |
MUTT: Metric Unit TesTing for Language Generation Tasks
| Title | MUTT: Metric Unit TesTing for Language Generation Tasks |
| Authors | William Boag, Renan Campos, Kate Saenko, Anna Rumshisky |
| Abstract | |
| Tasks | Image Captioning, Machine Translation, Text Generation, Video Captioning |
| Published | 2016-08-01 |
| URL | https://www.aclweb.org/anthology/P16-1182/ |
| https://www.aclweb.org/anthology/P16-1182 | |
| PWC | https://paperswithcode.com/paper/mutt-metric-unit-testing-for-language |
| Repo | |
| Framework | |
Dual Decomposed Learning with Factorwise Oracle for Structural SVM of Large Output Domain
| Title | Dual Decomposed Learning with Factorwise Oracle for Structural SVM of Large Output Domain |
| Authors | Ian En-Hsu Yen, Xiangru Huang, Kai Zhong, Ruohan Zhang, Pradeep K. Ravikumar, Inderjit S. Dhillon |
| Abstract | Many applications of machine learning involve structured output with large domain, where learning of structured predictor is prohibitive due to repetitive calls to expensive inference oracle. In this work, we show that, by decomposing training of Structural Support Vector Machine (SVM) into a series of multiclass SVM problems connected through messages, one can replace expensive structured oracle with Factorwise Maximization Oracle (FMO) that allows efficient implementation of complexity sublinear to the factor domain. A Greedy Direction Method of Multiplier (GDMM) algorithm is proposed to exploit sparsity of messages which guarantees $\epsilon$ sub-optimality after $O(log(1/\epsilon))$ passes of FMO calls. We conduct experiments on chain-structured problems and fully-connected problems of large output domains. The proposed approach is orders-of-magnitude faster than the state-of-the-art training algorithms for Structural SVM. |
| Tasks | |
| Published | 2016-12-01 |
| URL | http://papers.nips.cc/paper/6422-dual-decomposed-learning-with-factorwise-oracle-for-structural-svm-of-large-output-domain |
| http://papers.nips.cc/paper/6422-dual-decomposed-learning-with-factorwise-oracle-for-structural-svm-of-large-output-domain.pdf | |
| PWC | https://paperswithcode.com/paper/dual-decomposed-learning-with-factorwise |
| Repo | |
| Framework | |
Results of the WMT16 Tuning Shared Task
| Title | Results of the WMT16 Tuning Shared Task |
| Authors | Bushra Jawaid, Amir Kamran, Milo{\v{s}} Stanojevi{'c}, Ond{\v{r}}ej Bojar |
| Abstract | |
| Tasks | Machine Translation |
| Published | 2016-08-01 |
| URL | https://www.aclweb.org/anthology/W16-2303/ |
| https://www.aclweb.org/anthology/W16-2303 | |
| PWC | https://paperswithcode.com/paper/results-of-the-wmt16-tuning-shared-task |
| Repo | |
| Framework | |
Analysis of Policy Agendas: Lessons Learned from Automatic Topic Classification of Croatian Political Texts
| Title | Analysis of Policy Agendas: Lessons Learned from Automatic Topic Classification of Croatian Political Texts |
| Authors | Mladen Karan, Jan {\v{S}}najder, Daniela {\v{S}}irini{'c}, Goran Glava{\v{s}} |
| Abstract | |
| Tasks | Decision Making |
| Published | 2016-08-01 |
| URL | https://www.aclweb.org/anthology/W16-2102/ |
| https://www.aclweb.org/anthology/W16-2102 | |
| PWC | https://paperswithcode.com/paper/analysis-of-policy-agendas-lessons-learned |
| Repo | |
| Framework | |
Combinatorics vs Grammar: Archeology of Computational Poetry in Tape Mark I
| Title | Combinatorics vs Grammar: Archeology of Computational Poetry in Tape Mark I |
| Authors | Aless Mazzei, ro, Andrea Valle |
| Abstract | |
| Tasks | Text Generation |
| Published | 2016-09-01 |
| URL | https://www.aclweb.org/anthology/W16-5509/ |
| https://www.aclweb.org/anthology/W16-5509 | |
| PWC | https://paperswithcode.com/paper/combinatorics-vs-grammar-archeology-of |
| Repo | |
| Framework | |
Overview of NLP-TEA 2016 Shared Task for Chinese Grammatical Error Diagnosis
| Title | Overview of NLP-TEA 2016 Shared Task for Chinese Grammatical Error Diagnosis |
| Authors | Lung-Hao Lee, Gaoqi Rao, Liang-Chih Yu, Endong Xun, Baolin Zhang, Li-Ping Chang |
| Abstract | This paper presents the NLP-TEA 2016 shared task for Chinese grammatical error diagnosis which seeks to identify grammatical error types and their range of occurrence within sentences written by learners of Chinese as foreign language. We describe the task definition, data preparation, performance metrics, and evaluation results. Of the 15 teams registered for this shared task, 9 teams developed the system and submitted a total of 36 runs. We expected this evaluation campaign could lead to the development of more advanced NLP techniques for educational applications, especially for Chinese error detection. All data sets with gold standards and scoring scripts are made publicly available to researchers. |
| Tasks | Grammatical Error Correction |
| Published | 2016-12-01 |
| URL | https://www.aclweb.org/anthology/W16-4906/ |
| https://www.aclweb.org/anthology/W16-4906 | |
| PWC | https://paperswithcode.com/paper/overview-of-nlp-tea-2016-shared-task-for |
| Repo | |
| Framework | |
Visual Question Answering with Question Representation Update (QRU)
| Title | Visual Question Answering with Question Representation Update (QRU) |
| Authors | Ruiyu Li, Jiaya Jia |
| Abstract | Our method aims at reasoning over natural language questions and visual images. Given a natural language question about an image, our model updates the question representation iteratively by selecting image regions relevant to the query and learns to give the correct answer. Our model contains several reasoning layers, exploiting complex visual relations in the visual question answering (VQA) task. The proposed network is end-to-end trainable through back-propagation, where its weights are initialized using pre-trained convolutional neural network (CNN) and gated recurrent unit (GRU). Our method is evaluated on challenging datasets of COCO-QA and VQA and yields state-of-the-art performance. |
| Tasks | Question Answering, Visual Question Answering |
| Published | 2016-12-01 |
| URL | http://papers.nips.cc/paper/6261-visual-question-answering-with-question-representation-update-qru |
| http://papers.nips.cc/paper/6261-visual-question-answering-with-question-representation-update-qru.pdf | |
| PWC | https://paperswithcode.com/paper/visual-question-answering-with-question |
| Repo | |
| Framework | |
Merged bilingual trees based on Universal Dependencies in Machine Translation
| Title | Merged bilingual trees based on Universal Dependencies in Machine Translation |
| Authors | David Mare{\v{c}}ek |
| Abstract | |
| Tasks | Language Modelling, Machine Translation |
| Published | 2016-08-01 |
| URL | https://www.aclweb.org/anthology/W16-2318/ |
| https://www.aclweb.org/anthology/W16-2318 | |
| PWC | https://paperswithcode.com/paper/merged-bilingual-trees-based-on-universal |
| Repo | |
| Framework | |
LIMSI’s Contribution to the WMT’16 Biomedical Translation Task
| Title | LIMSI’s Contribution to the WMT’16 Biomedical Translation Task |
| Authors | Julia Ive, Aur{'e}lien Max, Fran{\c{c}}ois Yvon |
| Abstract | |
| Tasks | Machine Translation |
| Published | 2016-08-01 |
| URL | https://www.aclweb.org/anthology/W16-2337/ |
| https://www.aclweb.org/anthology/W16-2337 | |
| PWC | https://paperswithcode.com/paper/limsis-contribution-to-the-wmt16-biomedical |
| Repo | |
| Framework | |
Japanese Lexical Simplification for Non-Native Speakers
| Title | Japanese Lexical Simplification for Non-Native Speakers |
| Authors | Muhaimin Hading, Yuji Matsumoto, Maki Sakamoto |
| Abstract | This paper introduces Japanese lexical simplification. Japanese lexical simplification is the task of replacing difficult words in a given sentence to produce a new sentence with simple words without changing the original meaning of the sentence. We purpose a method of supervised regression learning to estimate difficulty ordering of words with statistical features obtained from two types of Japanese corpora. For the similarity of words, we use a Japanese thesaurus and dependency-based word embeddings. Evaluation of the proposed method is performed by comparing the difficulty ordering of the words. |
| Tasks | Language Modelling, Lexical Simplification, Machine Translation, Word Alignment, Word Embeddings |
| Published | 2016-12-01 |
| URL | https://www.aclweb.org/anthology/W16-4912/ |
| https://www.aclweb.org/anthology/W16-4912 | |
| PWC | https://paperswithcode.com/paper/japanese-lexical-simplification-for-non |
| Repo | |
| Framework | |
Investigating the Sources of Linguistic Alignment in Conversation
| Title | Investigating the Sources of Linguistic Alignment in Conversation |
| Authors | Gabriel Doyle, Michael C. Frank |
| Abstract | |
| Tasks | |
| Published | 2016-08-01 |
| URL | https://www.aclweb.org/anthology/P16-1050/ |
| https://www.aclweb.org/anthology/P16-1050 | |
| PWC | https://paperswithcode.com/paper/investigating-the-sources-of-linguistic |
| Repo | |
| Framework | |
Generating and Scoring Correction Candidates in Chinese Grammatical Error Diagnosis
| Title | Generating and Scoring Correction Candidates in Chinese Grammatical Error Diagnosis |
| Authors | Shao-Heng Chen, Yu-Lin Tsai, Chuan-Jie Lin |
| Abstract | Grammatical error diagnosis is an essential part in a language-learning tutoring system. Based on the data sets of Chinese grammar error detection tasks, we proposed a system which measures the likelihood of correction candidates generated by deleting or inserting characters or words, moving substrings to different positions, substituting prepositions with other prepositions, or substituting words with their synonyms or similar strings. Sentence likelihood is measured based on the frequencies of substrings from the space-removed version of Google n-grams. The evaluation on the training set shows that Missing-related and Selection-related candidate generation methods have promising performance. Our final system achieved a precision of 30.28{%} and a recall of 62.85{%} in the identification level evaluated on the test set. |
| Tasks | Decision Making |
| Published | 2016-12-01 |
| URL | https://www.aclweb.org/anthology/W16-4917/ |
| https://www.aclweb.org/anthology/W16-4917 | |
| PWC | https://paperswithcode.com/paper/generating-and-scoring-correction-candidates |
| Repo | |
| Framework | |
