Filters








5,991 Hits in 6.7 sec

A Survey on Multi-hop Question Answering and Generation [article]

Vaibhav Mavi
2022 arXiv   pre-print
The ability to answer multi-hop questions and perform multi step reasoning can significantly improve the utility of NLP systems.  ...  The notion of 'multiple hops' is somewhat abstract which results in a large variety of tasks that require multi-hop reasoning.  ...  [79] tackle the problem of difficulty controllable question generation (DQG) [44] by generating questions which require a particular number of reasoning hops.  ... 
arXiv:2204.09140v1 fatcat:tqr6472sjrfpvollvapwo3z3x4

Question-Aware Memory Network for Multi-hop Question Answering in Human-Robot Interaction [article]

Xinmeng Li, Mamoun Alazab, Qian Li, Keping Yu, Quanjun Yin
2021 arXiv   pre-print
We evaluate QA2MN on PathQuestion and WorldCup2014, two representative datasets for complex multi-hop question answering.  ...  Knowledge graph question answering is an important technology in intelligent human-robot interaction, which aims at automatically giving answer to human natural language question with the given knowledge  ...  For multi-hop question, Xu et al.  ... 
arXiv:2104.13173v1 fatcat:3v43jnf74rfyzi2x54df7vpk5i

Multi-hop Inference for Sentence-level TextGraphs: How Challenging is Meaningfully Combining Information for Science Question Answering?

Peter Jansen
2018 Proceedings of the Twelfth Workshop on Graph-Based Methods for Natural Language Processing (TextGraphs-12)  
judgements across knowledge graphs built from three freetext corpora (including study guides and Simple Wikipedia).  ...  Question Answering for complex questions is often modelled as a graph construction or traversal task, where a solver must build or traverse a graph of facts that answer and explain a given question.  ...  Information aggregation or "multi-hop" graph traversal has been shown to be extremely challenging, with QA solvers generally showing only modest performance benefits when aggregating information, and diminishing  ... 
doi:10.18653/v1/w18-1703 dblp:conf/textgraphs/Jansen18 fatcat:affa552ayfauvhkjxl3mhaxvt4

Few-shot Multi-hop Question Answering over Knowledge Base [article]

Meihao Fan, Lei Zhang, Siyao Xiao, Yuru Liang
2022 arXiv   pre-print
Besides, we propose a data generation strategy, which enables our model to generalize well from few training samples.  ...  KBQA is a task that requires to answer questions by using semantic structured information in knowledge base.  ...  Beam Search It is worth to notice that when extending multi-hop relations of the two type questions above, Query Path Generation module often suffers from the vast searching space.  ... 
arXiv:2112.11909v2 fatcat:6thbal5nureldcishdwrovoyae

Answering Any-hop Open-domain Questions with Iterative Document Reranking [article]

Ping Nie, Yuyu Zhang, Arun Ramamurthy, Le Song
2021 arXiv   pre-print
Existing approaches for open-domain question answering (QA) are typically designed for questions that require either single-hop or multi-hop reasoning, which make strong assumptions of the complexity of  ...  Our method consistently achieves performance comparable to or better than the state-of-the-art on both single-hop and multi-hop open-domain QA datasets, including Natural Questions Open, SQuAD Open, and  ...  To the best of our knowledge, we are the first to propose graph-based document reranking method for open-domain multi-hop QA. 2 OVERVIEW 2.1 Problem Definition Given a factoid question, the task  ... 
arXiv:2009.07465v5 fatcat:m3hosnzbwvdx3k3g7x7tlm5zhi

UHop: An Unrestricted-Hop Relation Extraction Framework for Knowledge-Based Question Answering [article]

Zi-Yuan Chen, Chih-Hung Chang, Yi-Pei Chen, Jijnasa Nayak, Lun-Wei Ku
2019 arXiv   pre-print
In related work, an exhaustive search from all one-hop relations, two-hop relations, and so on to the max-hop relations in the knowledge graph is necessary but expensive.  ...  In relation extraction for knowledge-based question answering, searching from one entity to another entity via a single relation is called "one hop".  ...  Figure 1 shows the difference between previous work and our model in the scenario of multi-hop KBQA task with an simplified knowledge graph and the question "Who published the novel adapted into A Study  ... 
arXiv:1904.01246v1 fatcat:ham6ahicdzc33cgx6ib442rp6u

KQA Pro: A Dataset with Explicit Compositional Programs for Complex Question Answering over Knowledge Base [article]

Shulin Cao, Jiaxin Shi, Liangming Pan, Lunyiu Nie, Yutong Xiang, Lei Hou, Juanzi Li, Bin He, Hanwang Zhang
2022 arXiv   pre-print
Complex question answering over knowledge base (Complex KBQA) is challenging because it requires various compositional reasoning capabilities, such as multi-hop inference, attribute comparison, set operation  ...  Our codes and datasets can be obtained from https://github.com/shijx12/KQAPro_Baselines.  ...  EmbedKGQA utilizes knowledge graph embedding to improve multi-hop reasoning.  ... 
arXiv:2007.03875v3 fatcat:otqbqpbsore7xjmz6rj6i5ncdq

Multi-hop Inference for Sentence-level TextGraphs: How Challenging is Meaningfully Combining Information for Science Question Answering? [article]

Peter Jansen
2018 arXiv   pre-print
judgments across knowledge graphs built from three free-text corpora (including study guides and Simple Wikipedia).  ...  Question Answering for complex questions is often modeled as a graph construction or traversal task, where a solver must build or traverse a graph of facts that answer and explain a given question.  ...  Information aggregation or "multi-hop" graph traversal has been shown to be extremely challenging, with QA solvers generally showing only modest performance benefits when aggregating information, and diminishing  ... 
arXiv:1805.11267v1 fatcat:mujrhpg6drdpbngz3e6cuuliiy

An Intelligent Question Answering System based on Power Knowledge Graph [article]

Yachen Tang, Haiyun Han, Xianmao Yu, Jing Zhao, Guangyi Liu, Longfei Wei
2021 arXiv   pre-print
This method thoroughly combined knowledge graph and graph computing characteristics, realized high-speed multi-hop knowledge correlation reasoning analysis in tremendous knowledge.  ...  This article introduced a domain knowledge graph using the graph database and graph computing technologies from massive heterogeneous data in electric power.  ...  In terms of graph traversal, the tested queries cover 10 single-hop and 190 multi-hop queries; in terms of restriction conditions, the tested queries cover 20 single-condition and 180 multi-condition queries  ... 
arXiv:2106.09013v1 fatcat:xw4hnpu4dzh3bbrd4s76kphcxy

GMH: A General Multi-hop Reasoning Model for KG Completion [article]

Yao Zhang, Hongru Liang, Adam Jatowt, Wenqiang Lei, Xin Wei, Ning Jiang, Zhenglu Yang
2021 arXiv   pre-print
To the best of our knowledge, there lacks a general framework that approaches multi-hop reasoning in mixed long-short distance reasoning scenarios.  ...  We argue that there are two key issues for a general multi-hop reasoning model: i) where to go, and ii) when to stop.  ...  Second, the global knowledge learnt from graph structure, which has been overlooked by the existing multi-hop reasoning models, is incorporated in GMH.  ... 
arXiv:2010.07620v3 fatcat:d56bqdmtcfeppdfoevlwjmruei

Cooperative Reasoning on Knowledge Graph and Corpus: A Multi-agentReinforcement Learning Approach [article]

Yunan Zhang, Xiang Cheng, Heting Gao, Chengxiang Zhai
2019 arXiv   pre-print
Each agent learns its skill to complete its own task, hopping on KG or select knowledge from the corpus, via maximizing the reward for correctly answering the question.  ...  However, previous methods suffer from the incompleteness of the knowledge graph, namely the interested link or entity that can be missing in the knowledge graph(explicit missing).  ...  Related Work We now review related work in the directions of knowledge graph reasoning, multi-agent reinforcement learning, and fusion of KB and texts for question answering.  ... 
arXiv:1912.02206v1 fatcat:l6oks5pidzfnxiavb3m37py5cm

GraphDialog: Integrating Graph Knowledge into End-to-End Task-Oriented Dialogue Systems [article]

Shiquan Yang, Rui Zhang, Sarah Erfani
2020 arXiv   pre-print
To exploit the relations between entities in KBs, the model combines multi-hop reasoning ability based on the graph structure.  ...  End-to-end task-oriented dialogue systems aim to generate system responses directly from plain text inputs.  ...  Acknowledgements We would like to thank Bayu Distiawan Trisedya for his insightful discussions and the valuable feedbacks from all anonymous reviewers.  ... 
arXiv:2010.01447v1 fatcat:ubmh2rloqfdo7h3gpi4oueyvn4

Neural, Symbolic and Neural-Symbolic Reasoning on Knowledge Graphs [article]

Jing Zhang, Bo Chen, Lingxi Zhang, Xirui Ke, Haipeng Ding
2021 arXiv   pre-print
We survey two specific reasoning tasks, knowledge graph completion and question answering on knowledge graphs, and explain them in a unified reasoning framework.  ...  Since knowledge graphs can be viewed as the discrete symbolic representations of knowledge, reasoning on knowledge graphs can naturally leverage the symbolic techniques.  ...  Symbolic Reasoning Symbolic reasoning aims at deducing general logic rules from the knowledge graphs.  ... 
arXiv:2010.05446v5 fatcat:tc6fowebkzbv7df3cjyhkcu6uq

Stabilizing Adversarial Invariance Induction from Divergence Minimization Perspective

Yusuke Iwasawa, Kei Akuzawa, Yutaka Matsuo
2020 Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence  
This paper presents an analysis of the reasons for the optimization difficulties and provides a better optimization procedure by rethinking AII from a divergence minimization perspective.  ...  Adversarial invariance induction (AII) is a generic and powerful framework for enforcing an invariance to nuisance attributes into neural network representations.  ...  Our contributions are summarized as flows: • We address the multiple semantic issue where a relation in knowledge graph has different meanings on multi-hop knowledge graph reasoning, which is an essential  ... 
doi:10.24963/ijcai.2020/267 dblp:conf/ijcai/WanP00H20 fatcat:zfhotcefyfgqljabsvtu6iwj3y

HoVer: A Dataset for Many-Hop Fact Extraction And Claim Verification [article]

Yichen Jiang, Shikha Bordia, Zheng Zhong, Charles Dognin, Maneesh Singh, Mohit Bansal
2020 arXiv   pre-print
In HoVer, the claims require evidence to be extracted from as many as four English Wikipedia articles and embody reasoning graphs of diverse shapes.  ...  many-hop reasoning to achieve strong results.  ...  WorldTree: A corpus of explanation graphs for elementary sci- ence questions supporting multi-hop inference.  ... 
arXiv:2011.03088v2 fatcat:wtwwko4bxbfuta3cnk3usxcdy4
« Previous Showing results 1 — 15 out of 5,991 results