Filters








98,200 Hits in 5.7 sec

Seeing is Knowing! Fact-based Visual Question Answering using Knowledge Graph Embeddings [article]

Kiran Ramnath, Mark Hasegawa-Johnson
2021 arXiv   pre-print
Fact-based Visual Question Answering (FVQA), a challenging variant of VQA, requires a QA-system to include facts from a diverse knowledge graph (KG) in its reasoning process to produce an answer.  ...  KG embeddings are shown to hold complementary information to word embeddings: a combination of both metrics permits performance comparable to SOTA methods in the standard answer retrieval task, and significantly  ...  SPO triple: : {Bus, AtLocation, Bus stop} Answer Source: Knowledge Base Answer: Bus stop Answer predicted: Wait place Question: What can we find in the place shown in this image?  ... 
arXiv:2012.15484v2 fatcat:iwp5zoabynhnxcuxubqfivo4ie

[Re] Improving Multi-hop Question Answering over Knowledge Graphs using Knowledge Base Embeddings

Jishnu Jaykumar P, Ashish Sardana
2021 Zenodo  
Exploring the effect of various knowledge graph embedding models in the Knowledge Graph Embedding module. 3. Exploring the effect of various transformer models in the Question Embedding module. 4.  ...  Question-Answering models were trained from scratch as no pre-trained models were available for our particular dataset.  ...  Moreover, MetaQA-KG-50 3- https://apoorvumang.github.io ReScience C 7.2 (#15) -P and Sardana 2021 [Re] Improving Multi-hop Question Answering over Knowledge Graphs using Knowledge Base Embeddings  ... 
doi:10.5281/zenodo.4834941 fatcat:uu33s5olqjckfk4brnpfqbcmy4

Seeing is Knowing! Fact-based Visual Question Answering using Knowledge Graph Embeddings [article]

Kiran Ramnath, Mark Hasegawa-Johnson
2020
Fact-based Visual Question Answering (FVQA), a challenging variant of VQA, requires a QA-system to include facts from a diverse knowledge graph (KG) in its reasoning process to produce an answer.  ...  KG embeddings are shown to hold complementary information to word embeddings: a combination of both metrics permits performance comparable to SOTA methods in the standard answer retrieval task, and significantly  ...  Improving multi-hop question answering ligence. over knowledge graphs using knowledge base em- beddings.  ... 
doi:10.48550/arxiv.2012.15484 fatcat:lpsipbjzvndf3kg6w2ag3dkasu

Improving Multi-hop Question Answering over Knowledge Graphs using Knowledge Base Embeddings

Apoorv Saxena, Aditay Tripathi, Partha Talukdar
2020 Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics   unpublished
Knowledge Graphs (KG) are multi-relational graphs consisting of entities as nodes and relations among them as typed edges.  ...  Goal of the Question Answering over KG (KGQA) task is to answer natural language queries posed over the KG.  ...  EmbedKGQA, on the other hand, uses Knowledge Graph Embeddings rather than a localized sub-graph to answer the question.  ... 
doi:10.18653/v1/2020.acl-main.412 fatcat:h3ndz7gjabfq5agkjf435mdmna

Vision–Language–Knowledge Co-Embedding for Visual Commonsense Reasoning

JaeYun Lee, Incheol Kim
2021 Sensors  
Therefore, we propose a novel Vision–Language–Knowledge Co-embedding (ViLaKC) model that extracts knowledge graphs relevant to the question from an external knowledge base, ConceptNet, and uses them together  ...  with the input image to answer the question.  ...  Evaluation of Knowledge Graph Embedding Methods The first experiment was conducted to verify the effectiveness of the GCN-based knowledge graph embedding method employed in the ViLaKC model.  ... 
doi:10.3390/s21092911 pmid:33919196 fatcat:ffoa2x6cindp3masddynz2qdmu

Relational Graph Representation Learning for Open-Domain Question Answering [article]

Salvatore Vivona, Kaveh Hassani
2019 arXiv   pre-print
We introduce a relational graph neural network with bi-directional attention mechanism and hierarchical representation learning for open-domain question answering task.  ...  Our model can learn contextual representation by jointly learning and updating the query, knowledge graph, and document representations.  ...  In this paper, we propose a relational GNN for open-domain question answering that learns contextual knowledge graph embeddings by jointly updating the embeddings from a knowledge graph and a set of linked  ... 
arXiv:1910.08249v1 fatcat:hboox75tynahrdfin7pdr4dofa

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
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  ...  In addition, we incorporate graph context information into knowledge graph embedding model to increase the ability to represent entities and relations.  ...  First, we exploit the graph context information in knowledge base by pre-training KG embedding model.  ... 
arXiv:2104.13173v1 fatcat:3v43jnf74rfyzi2x54df7vpk5i

Leveraging Domain Context for Question Answering Over Knowledge Graph

Peihao Tong, Qifan Zhang, Junjie Yao
2019 Data Science and Engineering  
With the growing availability of different knowledge graphs in a variety of domains, question answering over knowledge graph (KG-QA) becomes a prevalent information retrieval approach.  ...  The new approach is especially beneficial for specific knowledge graphs with complex questions.  ...  Answer Extraction To fully utilize the structure information of the knowledge graph, we employ the knowledge graph embedding method TransE [5] .  ... 
doi:10.1007/s41019-019-00109-w fatcat:eqymus5u2jg5pkllu6kldi3gk4

Dynamic Key-value Memory Enhanced Multi-step Graph Reasoning for Knowledge-based Visual Question Answering [article]

Mingxiao Li, Marie-Francine Moens
2022 arXiv   pre-print
Knowledge-based visual question answering (VQA) is a vision-language task that requires an agent to correctly answer image-related questions using knowledge that is not presented in the given image.  ...  Most existing knowledge-based VQA systems process knowledge and image information similarly and ignore the fact that the knowledge base (KB) contains complete information about a triplet, while the extracted  ...  Knowledge-Based VQA Knowledge-based VQA (KVQA) requires the model to use knowledge outside the image to answer questions correctly.  ... 
arXiv:2203.02985v1 fatcat:nzovyyeamzhxhiczdeharlrlw4

Dynamic Key-Value Memory Enhanced Multi-Step Graph Reasoning for Knowledge-Based Visual Question Answering

Mingxiao Li, Marie-Francine Moens
2022 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Knowledge-based visual question answering (VQA) is a vision-language task that requires an agent to correctly answer image-related questions using knowledge that is not presented in the given image.  ...  Most existing knowledge-based VQA systems process knowledge and image information similarly and ignore the fact that the knowledge base (KB) contains complete information about a triplet, while the extracted  ...  Knowledge-Based VQA Knowledge-based VQA (KVQA) requires the model to use knowledge outside the image to answer questions correctly.  ... 
doi:10.1609/aaai.v36i10.21346 fatcat:5mfq3vzq5vbqzpmj4qorih4wmi

Retrieve-then-extract Based Knowledge Graph Querying Using Graph Neural Networks [article]

Hanning Gao, Lingfei Wu, Po Hu, Zhihua Wei, Fangli Xu, Bo Long
2022 arXiv   pre-print
The abstract of Retrieve-then-extract Based Knowledge Graph Querying Using Graph Neural Networks will be updated here.  ...  RELATED WORK Knowledge Graph Question Answering With the rapid development of large-scale knowledge graphs (KG) such as DBpedia [2] and Freebase [7] , question answering over knowledge graph has attracted  ...  In general, Knowledge Graph Question Answering has two mainstream research methods, namely semantic parsing based methods and retrieve-then-extract methods. Semantic parsing based methods.  ... 
arXiv:2111.10541v3 fatcat:7hqfa5noe5bwxi5jyalhmvs4cm

Out of the Box: Reasoning with Graph Convolution Nets for Factual Visual Question Answering [article]

Medhini Narasimhan, Svetlana Lazebnik, Alexander G. Schwing
2018 arXiv   pre-print
Accurately answering a question about a given image requires combining observations with general knowledge.  ...  To advance research in this direction a novel 'fact-based' visual question answering (FVQA) task has been introduced recently along with a large set of curated facts which link two entities, i.e., two  ...  Acknowledgments: This material is based upon work supported in part by the National Science Foundation under Grant No. 1718221, Samsung, 3M, and the IBM-ILLINOIS Center for Cognitive Computing Systems  ... 
arXiv:1811.00538v1 fatcat:qzsd4jbcozg73e5vxerhmvfxmu

Improving Question Answering over Knowledge Graphs Using Graph Summarization [article]

Sirui Li, Kok Kai Wong, Dengya Zhu, Chun Che Fung
2022 arXiv   pre-print
Question Answering (QA) systems over Knowledge Graphs (KGs) (KGQA) automatically answer natural language questions using triples contained in a KG.  ...  Previous KGQAs have attempted to represent entities using Knowledge Graph Embedding (KGE) and Deep Learning (DL) methods.  ...  Instead of embedding the whole knowledge graph, Sun et al. [19, 20] extracted question-related subgraphs and then updated node embeddings by a single-layer GCN.  ... 
arXiv:2203.13570v1 fatcat:wdqxdlz6gzgrxktlau72osqrta

Variational Reasoning for Question Answering with Knowledge Graph [article]

Yuyu Zhang, Hanjun Dai, Zornitsa Kozareva, Alexander J. Smola, Le Song
2017 arXiv   pre-print
However, it is challenging to build QA systems which can learn to reason over knowledge graphs based on question-answer pairs alone.  ...  Second, many questions require multi-hop logic reasoning over the knowledge graph to retrieve the answers.  ...  Question answering with KG: Given a question q, the algorithm is asked to output an entity in the knowledge graph which properly answers the question.  ... 
arXiv:1709.04071v5 fatcat:54ccbhwfkzdz3ipwt6ordiwtr4

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.  ...  Specifically, we review the mechanisms of two typical KG reasoning tasks -knowledge graph completion (KGC) and knowledge graph question answering (KGQA).  ... 
arXiv:2010.05446v5 fatcat:tc6fowebkzbv7df3cjyhkcu6uq
« Previous Showing results 1 — 15 out of 98,200 results