Filters








28 Hits in 3.3 sec

Coarse-grain Fine-grain Coattention Network for Multi-evidence Question Answering [article]

Victor Zhong, Caiming Xiong, Nitish Shirish Keskar, Richard Socher
2019 arXiv   pre-print
In this work, we propose the Coarse-grain Fine-grain Coattention Network (CFC), a new question answering model that combines information from evidence across multiple documents.  ...  The CFC consists of a coarse-grain module that interprets documents with respect to the query then finds a relevant answer, and a fine-grain module which scores each candidate answer by comparing its occurrences  ...  ACKNOWLEDGEMENT The authors thank Luke Zettlemoyer for his feedback and advice and Sewon Min for her help in preprocessing the TriviaQA dataset.  ... 
arXiv:1901.00603v2 fatcat:lrj7qkfmtrbazhixas4x4eigrm

XTQA: Span-Level Explanations of the Textbook Question Answering [article]

Jie Ma, Jun Liu, Junjun Li, Qinghua Zheng, Qingyu Yin, Jianlong Zhou, Yi Huang
2020 arXiv   pre-print
Textbook Question Answering (TQA) is a task that one should answer a diagram/non-diagram question given a large multi-modal context consisting of abundant essays and diagrams.  ...  To address this issue, we devise a novel architecture towards span-level eXplanations of the TQA (XTQA) based on our proposed coarse-to-fine grained algorithm, which can provide not only the answers but  ...  We propose a coarse-to-fine grained algorithm to provide the evidence spans of questions for students.  ... 
arXiv:2011.12662v3 fatcat:owfnq3q6kfdlxpxoyd53g3stl4

MoCA: Incorporating Multi-stage Domain Pretraining and Cross-guided Multimodal Attention for Textbook Question Answering [article]

Fangzhi Xu, Qika Lin, Jun Liu, Lingling Zhang, Tianzhe Zhao, Qi Chai, Yudai Pan
2021 arXiv   pre-print
Textbook Question Answering (TQA) is a complex multimodal task to infer answers given large context descriptions and abundant diagrams.  ...  Compared with Visual Question Answering (VQA), TQA contains a large number of uncommon terminologies and various diagram inputs.  ...  is proposed to address multimodal fusion challenges. • XTQA (Ma et al. 2020): It designs a coarse-to-fine algo- rithm to generate span-level evidences. • RAFR (Ma et al. 2021): A fine-grained reasoning  ... 
arXiv:2112.02839v1 fatcat:waw2igbhozfvxfxnun2ncnobbq

Fine-tuning Multi-hop Question Answering with Hierarchical Graph Network [article]

Guanming Xiong
2020 arXiv   pre-print
In this paper, we present a two stage model for multi-hop question answering.  ...  The first stage is a hierarchical graph network, which is used to reason over multi-hop question and is capable to capture different levels of granularity using the nature structure(i.e., paragraphs, questions  ...  Coarse-grain fine-grain coattention network for multi-evidence question answering. In 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019.  ... 
arXiv:2004.13821v2 fatcat:cl5z74cbzze2rkwjvqajhhugwe

Multi-hop Reading Comprehension across Multiple Documents by Reasoning over Heterogeneous Graphs [article]

Ming Tu, Guangtao Wang, Jing Huang, Yun Tang, Xiaodong He, Bowen Zhou
2019 arXiv   pre-print
We employ Graph Neural Networks (GNN) based message passing algorithms to accumulate evidences on the proposed HDE graph.  ...  Multi-hop reading comprehension (RC) across documents poses new challenge over single-document RC because it requires reasoning over multiple documents to reach the final answer.  ...  Acknowledgements We would like to thank Johannes Welbl for running evaluation on our submitted model.  ... 
arXiv:1905.07374v2 fatcat:epznpgcp7vgovminxdcn53t5yq

The Natural Language Decathlon: Multitask Learning as Question Answering [article]

Bryan McCann and Nitish Shirish Keskar and Caiming Xiong and Richard Socher
2018 arXiv   pre-print
We cast all tasks as question answering over a context.  ...  Furthermore, we present a new Multitask Question Answering Network (MQAN) jointly learns all tasks in decaNLP without any task-specific modules or parameters in the multitask setting.  ...  The multitask question answering network (MQAN) is designed for decaNLP and makes use of a novel dual coattention and multi-pointer-generator decoder to multitask across all tasks in decaNLP.  ... 
arXiv:1806.08730v1 fatcat:pdvwr3fqfrdnjdzwotzahsjf3e

Smarnet: Teaching Machines to Read and Comprehend Like Human [article]

Zheqian Chen, Rongqin Yang, Bin Cao, Zhou Zhao, Deng Cai, Xiaofei He
2017 arXiv   pre-print
We then guide the machines to read in an interactive way with attention mechanism and memory network. Finally we add a checking layer to refine the answer for insurance.  ...  Machine Comprehension (MC) is a challenging task in Natural Language Processing field, which aims to guide the machine to comprehend a passage and answer the given question.  ...  Therefore, we propose a multi-hop memory network which allows to reread the question and answer.  ... 
arXiv:1710.02772v1 fatcat:hkdqwqvxtzecdpti5jpl2cogfa

Multi-hop Reading Comprehension across Multiple Documents by Reasoning over Heterogeneous Graphs

Ming Tu, Guangtao Wang, Jing Huang, Yun Tang, Xiaodong He, Bowen Zhou
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics  
We employ Graph Neural Networks (GNN) based message passing algorithms to accumulate evidences on the proposed HDE graph.  ...  Multi-hop reading comprehension (RC) across documents poses new challenge over singledocument RC because it requires reasoning over multiple documents to reach the final answer.  ...  Acknowledgements We would like to thank Johannes Welbl from University College London for running evaluation on our submitted model.  ... 
doi:10.18653/v1/p19-1260 dblp:conf/acl/TuWHTHZ19 fatcat:os2vkhkzynh5bl2xk44yzhdpkq

Complex Factoid Question Answering with a Free-Text Knowledge Graph

Chen Zhao, Chenyan Xiong, Xin Qian, Jordan Boyd-Graber
2020 Proceedings of The Web Conference 2020  
Experiments on three question answering datasets show DELFT can answer entity-rich questions better than machine reading based models, bert-based answer ranking and memory networks.  ...  A novel graph neural network reasons over the free-text graph-combining evidence on the nodes via information along edge sentences-to select a final answer.  ...  Fine-grain Coattention Network for Multi-evidence Question An- and Quoc V Le. 2019.  ... 
doi:10.1145/3366423.3380197 dblp:conf/www/ZhaoXQB20 fatcat:ahvzga5qdjdfph36gpkuikn4oe

Ask to Understand: Question Generation for Multi-hop Question Answering [article]

Jiawei Li, Mucheng Ren, Yang Gao, Yizhe Yang
2022 arXiv   pre-print
Multi-hop Question Answering (QA) requires the machine to answer complex questions by finding scattering clues and reasoning from multiple documents.  ...  Graph Network (GN) and Question Decomposition (QD) are two common approaches at present.  ...  Related Work Multi-hop QA In multi-hop QA, the evidence for reasoning answers is scattered across multiple sentences.  ... 
arXiv:2203.09073v1 fatcat:3i34k5linfbfxppvv5mu7ub5g4

Explore, Propose, and Assemble: An Interpretable Model for Multi-Hop Reading Comprehension

Yichen Jiang, Nitish Joshi, Yen-Chun Chen, Mohit Bansal
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics  
Intuitively, EPAr approximates the coarse-to-fine-grained comprehension behavior of human readers when facing multiple long documents.  ...  Multi-hop reading comprehension requires the model to explore and connect relevant information from multiple sentences/documents in order to answer the question about the context.  ...  Acknowledgement We would like to thank Johannes Welbl for helping test our system on WikiHop and MedHop. We thank the reviewers for their helpful comments.  ... 
doi:10.18653/v1/p19-1261 dblp:conf/acl/JiangJCB19 fatcat:wxedrdchgzgehbybckxisks76e

Cross-Modal Commentator: Automatic Machine Commenting Based on Cross-Modal Information

Pengcheng Yang, Zhihan Zhang, Fuli Luo, Lei Li, Chengyang Huang, Xu Sun
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics  
For instance, graphic news contains plenty of images in addition to text.  ...  We construct a largescale dataset for this task and explore several representative methods.  ...  Acknowledgement We thank the anonymous reviewers for their thoughtful comments. Xu Sun is the contact author of this paper.  ... 
doi:10.18653/v1/p19-1257 dblp:conf/acl/YangZLLHS19 fatcat:jfvat3ij5bg25fgwrfo6san7me

Question-Driven Purchasing Propensity Analysis for Recommendation

Long Chen, Ziyu Guan, Qibin Xu, Qiong Zhang, Huan Sun, Guangyue Lu, Deng Cai
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
To address this recommendation problem, we propose a novel Question-Driven Attentive Neural Network (QDANN) to assess the instant demands of questioners and the eligibility of products based on user generated  ...  The attention mechanisms can be used to provide explanations for recommendations. We evaluate QDANN in three domains of Taobao.  ...  (Chen et al. 2019 ) proposed a multi-task attentive network for plausible answer identification from reviews.  ... 
doi:10.1609/aaai.v34i01.5331 fatcat:aipdzryhybexhcmmb4yvfzwuui

New Ideas and Trends in Deep Multimodal Content Understanding: A Review

Wei Chen, Weiping Wang, Li Liu, Michael S. Lew
2020 Neurocomputing  
These models go beyond the simple image classifiers in which they can do uni-directional (e.g. image captioning, image generation) and bi-directional (e.g. cross-modal retrieval, visual question answering  ...  Finally, we include several promising directions for future research.  ...  For example, some fine-grained attributes including texture, shape and color can be specified during deep network training.  ... 
doi:10.1016/j.neucom.2020.10.042 fatcat:hyjkj5enozfrvgzxy6avtbmoxu

New Ideas and Trends in Deep Multimodal Content Understanding: A Review [article]

Wei Chen and Weiping Wang and Li Liu and Michael S. Lew
2020 arXiv   pre-print
These models go beyond the simple image classifiers in which they can do uni-directional (e.g. image captioning, image generation) and bi-directional (e.g. cross-modal retrieval, visual question answering  ...  Finally, we include several promising directions for future research.  ...  For example, some fine-grained attributes including texture, shape and color can be specified during deep network training.  ... 
arXiv:2010.08189v1 fatcat:2l7molbcn5hf3oyhe3l52tdwra
« Previous Showing results 1 — 15 out of 28 results