Filters








653 Hits in 4.2 sec

End-to-End Trainable Attentive Decoder for Hierarchical Entity Classification

Sanjeev Karn, Ulli Waltinger, Hinrich Schütze
2017 Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 2, Short Papers  
We address fine-grained entity classification and propose a novel attention-based recurrent neural network (RNN) encoderdecoder that generates paths in the type hierarchy and can be trained end-to-end.  ...  We show that our model performs better on fine-grained entity classification than prior work that relies on flat or local classifiers that do not directly model hierarchical structure.  ...  We thank Stephan Baier, Siemens CT members and the anonymous reviewers for valuable feedback. This research was supported by Bundeswirtschaftsministerium (bmwi. de), grant 01MD15010A (Smart Data Web).  ... 
doi:10.18653/v1/e17-2119 dblp:conf/eacl/SchutzeWK17 fatcat:frulzitxgrdsvbaohokyxwtxf4

Hierarchical Attention and Knowledge Matching Networks With Information Enhancement for End-to-End Task-Oriented Dialog Systems

Junqing He, Bing Wang, Mingming Fu, Tianqi Yang, Xuemin Zhao
2019 IEEE Access  
Nowadays, most end-to-end task-oriented dialog systems are based on sequence-to-sequence (Seq2seq), which is an encoder-decoder framework.  ...  To solve this problem, we introduce hierarchical attention and knowledge matching networks with information enhancement (HAMI) for task-oriented dialog systems.  ...  The whole system is end-to-end trainable. The framework of our system is illustrated in Figure 1 . A.  ... 
doi:10.1109/access.2019.2892730 fatcat:rt2d4j77xrhmtlmzg6a42utruu

Story Ending Generation with Incremental Encoding and Commonsense Knowledge

Jian Guan, Yansen Wang, Minlie Huang
2019 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
In addition, commonsense knowledge is applied through multi-source attention to facilitate story comprehension, and thus to help generate coherent and reasonable endings.  ...  Generating a reasonable ending for a given story context, i.e., story ending generation, is a strong indication of story comprehension.  ...  We would like to thank Prof. Xiaoyan Zhu for her generous support.  ... 
doi:10.1609/aaai.v33i01.33016473 fatcat:thvza6vddvafvjerc7dzcom4dq

Story Ending Generation with Incremental Encoding and Commonsense Knowledge [article]

Jian Guan, Yansen Wang, Minlie Huang
2018 arXiv   pre-print
In addition, commonsense knowledge is applied through multi-source attention to facilitate story comprehension, and thus to help generate coherent and reasonable endings.  ...  Generating a reasonable ending for a given story context, i.e., story ending generation, is a strong indication of story comprehension.  ...  We would like to thank Prof. Xiaoyan Zhu for her generous support.  ... 
arXiv:1808.10113v3 fatcat:ydhklq5qkzdkflrfircl4tmhim

Visual Relationship Detection Using Part-and-Sum Transformers with Composite Queries [article]

Qi Dong, Zhuowen Tu, Haofu Liao, Yuting Zhang, Vijay Mahadevan, Stefano Soatto
2021 arXiv   pre-print
In this paper, we present a new approach, denoted Part-and-Sum detection Transformer (PST), to perform end-to-end visual composite set detection.  ...  be detected in a hierarchical fashion.  ...  Algorithms performing inference hierarchically exist [57, 34, 38] but they are not trained end-to-end.  ... 
arXiv:2105.02170v2 fatcat:z5dn6rrbxffkditvnh6icazkra

Deep Neural Networks for Relation Extraction [article]

Tapas Nayak
2021 arXiv   pre-print
Finally, we propose a hierarchical entity graph convolutional network for relation extraction across documents.  ...  In this thesis, we first propose a syntax-focused multi-factor attention network model for finding the relation between two entities.  ...  This decoder consists of two pointer networks which find the start and end location of the two entities in a sentence, and a classification net-work which identifies the relation between them.  ... 
arXiv:2104.01799v1 fatcat:vmatz7gxazd4xnm2oprncd5mm4

A Survey on Dialogue Systems: Recent Advances and New Frontiers [article]

Hongshen Chen, Xiaorui Liu, Dawei Yin, Jiliang Tang
2018 arXiv   pre-print
Dialogue systems have attracted more and more attention.  ...  For dialogue systems, deep learning can leverage a massive amount of data to learn meaningful feature representations and response generation strategies, while requiring a minimum amount of hand-crafting  ...  With the advance of end-to-end neural generative models in recent years, many attempts have been made to construct an end-to-end trainable framework for task-oriented dialogue systems.  ... 
arXiv:1711.01731v3 fatcat:6wuovcynqbhlzmuorchn4mn6ma

Learn to Focus: Hierarchical Dynamic Copy Network for Dialogue State Tracking [article]

Linhao Zhang, Houfeng Wang
2021 arXiv   pre-print
Based on the encoder-decoder framework, we adopt a hierarchical copy approach that calculates two levels of attention at the word- and turn-level, which are then renormalized to obtain the final copy distribution  ...  A focus loss term is employed to encourage the model to assign the highest turn-level attention weight to the most informative turn.  ...  Hierarchical Attention Network -Our hierarchical dynamic copy mechanism draws inspiration from Hierarchical Attention Network (HAN) (Yang et al., 2016) , which was first proposed for text classification  ... 
arXiv:2107.11778v1 fatcat:5wxiswur7beunhfum4okukrrpm

Iterative-Deepening Conflict-Based Search

Eli Boyarski, Ariel Felner, Daniel Harabor, Peter J. Stuckey, Liron Cohen, Jiaoyang Li, Sven Koenig
2020 Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence  
Conflict-Based Search (CBS) is a leading algorithm for optimal Multi-Agent Path Finding (MAPF). CBS variants typically compute MAPF solutions using some form of A* search.  ...  However, they often do so under strict time limits so as to avoid exhausting the available memory.  ...  We thank all anonymous reviewers for their constructive comments.  ... 
doi:10.24963/ijcai.2020/561 dblp:conf/ijcai/YuanZPZSG20 fatcat:j7ygfwlc4zgybnjhuqf243dsa4

Fast Abstractive Summarization with Reinforce-Selected Sentence Rewriting

Yen-Chun Chen, Mohit Bansal
2018 Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)  
We use a novel sentence-level policy gradient method to bridge the nondifferentiable computation between these two neural networks in a hierarchical way, while maintaining language fluency.  ...  ., compresses and paraphrases) to generate a concise overall summary.  ...  Acknowledgments We thank the anonymous reviewers for their helpful comments.  ... 
doi:10.18653/v1/p18-1063 dblp:conf/acl/BansalC18 fatcat:7zhltuzqq5frdfhnkzyjmmdhia

Phrase-based Image Captioning with Hierarchical LSTM Model [article]

Ying Hua Tan, Chee Seng Chan
2017 arXiv   pre-print
In this paper, we propose a phrase-based hierarchical Long Short-Term Memory (phi-LSTM) model to generate image description.  ...  It consists of a phrase decoder at the bottom hierarchy to decode noun phrases of variable length, and an abbreviated sentence decoder at the upper hierarchy to decode an abbreviated form of the image  ...  To this end, we design a phrase-based hierarchical LSTM model, namely phi-LSTM that consists of a phrase decoder and an abbreviated sentence (AS) decoder to generate image description from phrase to arXiv  ... 
arXiv:1711.05557v1 fatcat:slhffic5tvgxrh3tftpfe7c4ju

Conversational Question Answering over Knowledge Graphs with Transformer and Graph Attention Networks [article]

Endri Kacupaj, Joan Plepi, Kuldeep Singh, Harsh Thakkar, Jens Lehmann, Maria Maleshkova
2021 arXiv   pre-print
LASAGNE uses a transformer model for generating the base logical forms, while the Graph Attention model is used to exploit correlations between (entity) types and predicates to produce node representations  ...  For this task, we propose LASAGNE (muLti-task semAntic parSing with trAnsformer and Graph atteNtion nEtworks).  ...  Acknowledgments The project leading to this publication has received funding from the European Union's Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No.  ... 
arXiv:2104.01569v2 fatcat:nksqgjhp45cpbck56i6ld6bgrm

Learning to Memorize in Neural Task-Oriented Dialogue Systems [article]

Chien-Sheng Wu
2019 arXiv   pre-print
We also propose a recorded delexicalization copy strategy to replace real entity values with ordered entity types.  ...  Mem2Seq is the first model to combine multi-hop memory attention with the idea of the copy mechanism. GLMP further introduces the concept of response sketching and double pointers copying.  ...  Short Summary We present an end-to-end trainable memory-to-sequence (Mem2Seq) model for task-oriented dialogue systems.  ... 
arXiv:1905.07687v1 fatcat:upvqqczyajdtnf45bcszxs6nai

Fast Abstractive Summarization with Reinforce-Selected Sentence Rewriting [article]

Yen-Chun Chen, Mohit Bansal
2018 arXiv   pre-print
We use a novel sentence-level policy gradient method to bridge the non-differentiable computation between these two neural networks in a hierarchical way, while maintaining language fluency.  ...  ., compresses and paraphrases) to generate a concise overall summary.  ...  Acknowledgments We thank the anonymous reviewers for their helpful comments.  ... 
arXiv:1805.11080v1 fatcat:4w3iawlhp5fhvp2yclghygsr2e

TIMME: Twitter Ideology-detection via Multi-task Multi-relational Embedding [article]

Zhiping Xiao, Weiping Song, Haoyan Xu, Zhicheng Ren, Yizhou Sun
2020 arXiv   pre-print
Our findings include: links can lead to good classification outcomes without text; conservative voice is under-represented on Twitter; follow is the most important relation to predict ideology; retweet  ...  Experimental results showed that TIMME is overall better than the state-of-the-art models for ideology detection on Twitter.  ...  Figure 3 : The two types of decoder in our multi-task framework, referred to as TIMME and TIMME-hierarchical.  ... 
arXiv:2006.01321v2 fatcat:efvsu7r2ijgp3pab5c3bwubvxu
« Previous Showing results 1 — 15 out of 653 results