{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,5,6]],"date-time":"2025-05-06T05:49:48Z","timestamp":1746510588710},"reference-count":97,"publisher":"MIT Press","issue":"4","license":[{"start":{"date-parts":[[2021,8,5]],"date-time":"2021-08-05T00:00:00Z","timestamp":1628121600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2021,12,23]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Currently, deep learning models have been widely adopted and achieved promising results on various application domains. Despite their intriguing performance, most deep learning models function as black boxes, lacking explicit reasoning capabilities and explanations, which are usually essential for complex problems. Take joint inference in information extraction as an example. This task requires the identification of multiple structured knowledge from texts, which is inter-correlated, including entities, events, and the relationships between them. Various deep neural networks have been proposed to jointly perform entity extraction and relation prediction, which only propagate information implicitly via representation learning. However, they fail to encode the intensive correlations between entity types and relations to enforce their coexistence. On the other hand, some approaches adopt rules to explicitly constrain certain relational facts, although the separation of rules with representation learning usually restrains the approaches with error propagation. Moreover, the predefined rules are inflexible and might result in negative effects when data is noisy. To address these limitations, we propose a variational deep logic network that incorporates both representation learning and relational reasoning via the variational EM algorithm. The model consists of a deep neural network to learn high-level features with implicit interactions via the self-attention mechanism and a relational logic network to explicitly exploit target interactions. These two components are trained interactively to bring the best of both worlds. We conduct extensive experiments ranging from fine-grained sentiment terms extraction, end-to-end relation prediction, to end-to-end event extraction to demonstrate the effectiveness of our proposed method.<\/jats:p>","DOI":"10.1162\/coli_a_00415","type":"journal-article","created":{"date-parts":[[2021,8,5]],"date-time":"2021-08-05T14:36:01Z","timestamp":1628174161000},"page":"775-812","update-policy":"http:\/\/dx.doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":2,"title":["Variational Deep Logic Network for Joint Inference of Entities and Relations"],"prefix":"10.1162","volume":"47","author":[{"given":"Wenya","family":"Wang","sequence":"first","affiliation":[{"name":"School of Computer Science and Engineering, Nanyang Technological University, Singapore. wangwy@ntu.edu.sg"}]},{"given":"Sinno Jialin","family":"Pan","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Nanyang Technological University, Singapore. sinnopan@ntu.edu.sg"}]}],"member":"281","published-online":{"date-parts":[[2021,12,23]]},"reference":[{"key":"2022010319055114200_bib1","doi-asserted-by":"publisher","first-page":"1723","DOI":"10.18653\/v1\/D17-1181","article-title":"Global normalization of convolutional neural networks for joint entity and relation classification","volume-title":"EMNLP","author":"Adel","year":"2017"},{"key":"2022010319055114200_bib2","doi-asserted-by":"publisher","first-page":"34","DOI":"10.1016\/j.eswa.2018.07.032","article-title":"Joint entity recognition and relation extraction as a multi-head selection problem","volume":"114","author":"Bekoulis","year":"2018","journal-title":"Expert Systems with Applications"},{"key":"2022010319055114200_bib3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.18653\/v1\/D18-1307","article-title":"Adversarial training for multi-context joint entity and relation extraction","volume-title":"EMNLP","author":"Bekoulis","year":"2018"},{"key":"2022010319055114200_bib4","article-title":"Logical rule induction and theory learning using neural theorem proving","author":"Campero","year":"2018","journal-title":"CoRR"},{"key":"2022010319055114200_bib5","first-page":"551","article-title":"Exploiting syntactico-semantic structures for relation extraction","volume-title":"ACL-HLT","author":"Chan","year":"2011"},{"key":"2022010319055114200_bib6","doi-asserted-by":"publisher","first-page":"3685","DOI":"10.18653\/v1\/2020.acl-main.340","article-title":"Relation-aware collaborative learning for unified aspect-based sentiment analysis","volume-title":"Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics","author":"Chen","year":"2020"},{"key":"2022010319055114200_bib7","doi-asserted-by":"publisher","first-page":"3658","DOI":"10.1609\/aaai.v34i04.5774","article-title":"A constraint-based approach to learning and explanation","volume-title":"AAAI","author":"Ciravegna","year":"2020"},{"key":"2022010319055114200_bib8","doi-asserted-by":"publisher","first-page":"6300","DOI":"10.1609\/aaai.v33i01.33016300","article-title":"Joint extraction of entities and overlapping relations using position-attentive sequence labeling","volume-title":"AAAI","author":"Dai","year":"2019"},{"key":"2022010319055114200_bib9","first-page":"2702","article-title":"Discriminative embeddings of latent variable models for structured data","volume-title":"ICML","author":"Dai","year":"2016"},{"key":"2022010319055114200_bib10","doi-asserted-by":"publisher","first-page":"2815","DOI":"10.1007\/s11432-018-9801-4","article-title":"Bridging machine learning and logical reasoning by abductive learning","volume-title":"NeurIPS","author":"Dai","year":"2019"},{"issue":"4","key":"2022010319055114200_bib11","first-page":"611","article-title":"Neural-symbolic computing: An effective methodology for principled integration of machine learning and reasoning","volume":"6","author":"d\u2019Avila Garcez","year":"2019","journal-title":"FLAP"},{"key":"2022010319055114200_bib12","first-page":"4171","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","volume-title":"NAACL","author":"Devlin","year":"2019"},{"key":"2022010319055114200_bib13","doi-asserted-by":"publisher","first-page":"5308","DOI":"10.18653\/v1\/P19-1525","article-title":"Span-level model for relation extraction","volume-title":"ACL","author":"Dixit","year":"2019"},{"key":"2022010319055114200_bib14","article-title":"The automatic content extraction (ACE) program \u2013 tasks, data, and evaluation","volume-title":"LREC","author":"Doddington","year":"2004"},{"key":"2022010319055114200_bib15","article-title":"Neural logic machines","volume-title":"ICLR","author":"Dong","year":"2019"},{"key":"2022010319055114200_bib16","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1613\/jair.5714","article-title":"Learning explanatory rules from noisy data","volume":"61","author":"Evans","year":"2018","journal-title":"Journal of Artificial Intelligence Research"},{"issue":"1","key":"2022010319055114200_bib17","doi-asserted-by":"publisher","first-page":"81","DOI":"10.1007\/s10994-013-5392-1","article-title":"Fast relational learning using bottom clause propositionalization with artificial neural networks","volume":"94","author":"Fran\u00e7a","year":"2014","journal-title":"Machine Learning"},{"key":"2022010319055114200_bib18","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4684-3384-5","article-title":"Logic and data bases, symposium on logic and data bases, Centre D\u2019\u00e9tudes et de Recherches de Toulouse, 1977","volume-title":"Advances in Data Base Theory","author":"Gallaire","year":"1978"},{"key":"2022010319055114200_bib19","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4471-0211-3","volume-title":"Neural-Symbolic Learning Systems - Foundations and Applications","author":"Garcez","year":"2002"},{"key":"2022010319055114200_bib20","doi-asserted-by":"publisher","first-page":"192","DOI":"10.18653\/v1\/D16-1019","article-title":"Jointly embedding knowledge graphs and logical rules","volume-title":"EMNLP","author":"Guo","year":"2016"},{"key":"2022010319055114200_bib21","first-page":"2537","article-title":"Table filling multi-task recurrent neural network for joint entity and relation extraction","volume-title":"COLING","author":"Gupta","year":"2016"},{"key":"2022010319055114200_bib22","first-page":"1127","article-title":"Using cross-entity inference to improve event extraction","volume-title":"ACL-HLT","author":"Hong","year":"2011"},{"key":"2022010319055114200_bib23","doi-asserted-by":"publisher","first-page":"168","DOI":"10.1145\/1014052.1014073","article-title":"Mining and summarizing customer reviews","volume-title":"KDD","author":"Hu","year":"2004"},{"key":"2022010319055114200_bib24","doi-asserted-by":"publisher","first-page":"2410","DOI":"10.18653\/v1\/P16-1228","article-title":"Harnessing deep neural networks with logic rules","volume-title":"ACL","author":"Hu","year":"2016"},{"key":"2022010319055114200_bib25","first-page":"254","article-title":"Refining event extraction through cross-document inference","volume-title":"Proceedings of ACL-08: HLT","author":"Ji","year":"2008"},{"key":"2022010319055114200_bib26","first-page":"2279","article-title":"Incremental global event extraction","volume-title":"COLING","author":"Judea","year":"2016"},{"key":"2022010319055114200_bib27","first-page":"203","article-title":"Joint entity and relation extraction using card-pyramid parsing","volume-title":"CoNLL","author":"Kate","year":"2010"},{"key":"2022010319055114200_bib28","doi-asserted-by":"publisher","first-page":"917","DOI":"10.18653\/v1\/P17-1085","article-title":"Going out on a limb: Joint extraction of entity mentions and relations without dependency trees","volume-title":"ACL","author":"Katiyar","year":"2017"},{"key":"2022010319055114200_bib29","article-title":"Semi-supervised classification with graph convolutional networks","volume-title":"Proceedings of the 5th International Conference on Learning Representations","author":"Kipf","year":"2017"},{"key":"2022010319055114200_bib30","volume-title":"Triangular Norms","author":"Klement","year":"2013"},{"key":"2022010319055114200_bib31","first-page":"282","article-title":"Conditional random fields: Probabilistic models for segmenting and labeling sequence data","volume-title":"ICML","author":"Lafferty","year":"2001"},{"key":"2022010319055114200_bib32","doi-asserted-by":"publisher","first-page":"4877","DOI":"10.24963\/ijcai.2020\/679","article-title":"Graph neural networks meet neural-symbolic computing: A survey and perspective","volume-title":"IJCAI","author":"Lamb","year":"2020"},{"key":"2022010319055114200_bib33","first-page":"653","article-title":"Structure-aware review mining and summarization","volume-title":"COLING","author":"Li","year":"2010"},{"key":"2022010319055114200_bib34","first-page":"402","article-title":"Incremental joint extraction of entity mentions and relations","volume-title":"ACL","author":"Li","year":"2014"},{"key":"2022010319055114200_bib35","doi-asserted-by":"publisher","first-page":"1846","DOI":"10.3115\/v1\/D14-1198","article-title":"Constructing information networks using one single model","volume-title":"EMNLP","author":"Li","year":"2014"},{"key":"2022010319055114200_bib36","first-page":"292","article-title":"Augmenting neural networks with first-order logic","volume-title":"ACL","author":"Li","year":"2019"},{"key":"2022010319055114200_bib37","doi-asserted-by":"publisher","first-page":"1340","DOI":"10.18653\/v1\/P19-1129","article-title":"Entity-relation extraction as multi-turn question answering","volume-title":"ACL","author":"Li","year":"2019"},{"key":"2022010319055114200_bib38","doi-asserted-by":"publisher","first-page":"2886","DOI":"10.18653\/v1\/D17-1310","article-title":"Deep multi-task learning for aspect term extraction with memory interaction","volume-title":"EMNLP","author":"Li","year":"2017"},{"key":"2022010319055114200_bib39","first-page":"789","article-title":"Using document level cross-event inference to improve event extraction","volume-title":"ACL","author":"Liao","year":"2010"},{"key":"2022010319055114200_bib40","doi-asserted-by":"publisher","first-page":"2124","DOI":"10.18653\/v1\/P16-1200","article-title":"Neural relation extraction with selective attention over instances","volume-title":"ACL","author":"Lin","year":"2016"},{"key":"2022010319055114200_bib41","doi-asserted-by":"publisher","first-page":"7999","DOI":"10.18653\/v1\/2020.acl-main.713","article-title":"A joint neural model for information extraction with global features","volume-title":"Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics","author":"Lin","year":"2020"},{"key":"2022010319055114200_bib42","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D15-1168","article-title":"Fine-grained opinion mining with recurrent neural networks and word embeddings","volume-title":"EMNLP","author":"Liu","year":"2015"},{"key":"2022010319055114200_bib43","doi-asserted-by":"publisher","first-page":"1247","DOI":"10.18653\/v1\/D18-1156","article-title":"Jointly multiple events extraction via attention-based graph information aggregation","volume-title":"EMNLP","author":"Liu","year":"2018"},{"key":"2022010319055114200_bib44","doi-asserted-by":"publisher","first-page":"3036","DOI":"10.18653\/v1\/N19-1308","article-title":"A general framework for information extraction using dynamic span graphs","volume-title":"NAACL","author":"Luan","year":"2019"},{"key":"2022010319055114200_bib45","first-page":"3749","article-title":"DeepProbLog: Neural probabilistic logic programming","volume-title":"NeurIPS","author":"Manhaeve","year":"2018"},{"key":"2022010319055114200_bib46","doi-asserted-by":"publisher","first-page":"55","DOI":"10.3115\/v1\/P14-5010","article-title":"The Stanford CoreNLP natural language processing toolkit","volume-title":"ACL (System Demonstrations)","author":"Manning","year":"2014"},{"key":"2022010319055114200_bib47","first-page":"1614","article-title":"From softmax to sparsemax: A sparse model of attention and multi-label classification","volume-title":"ICML","author":"Martins","year":"2016"},{"key":"2022010319055114200_bib48","first-page":"1626","article-title":"Event extraction as dependency parsing","volume-title":"ACL-HLT","author":"McClosky","year":"2011"},{"key":"2022010319055114200_bib49","doi-asserted-by":"publisher","first-page":"5182","DOI":"10.1609\/aaai.v34i04.5962","article-title":"Differentiable reasoning on large knowledge bases and natural language","volume-title":"AAAI","author":"Minervini","year":"2020"},{"key":"2022010319055114200_bib50","article-title":"Adversarial sets for regularised neural link predictors","volume-title":"UAI","author":"Minervini","year":"2017"},{"key":"2022010319055114200_bib51","doi-asserted-by":"publisher","first-page":"65","DOI":"10.18653\/v1\/K18-1007","article-title":"Adversarially regularising neural NLI models to integrate logical background knowledge","volume-title":"CoNLL","author":"Minervini","year":"2018"},{"key":"2022010319055114200_bib52","doi-asserted-by":"publisher","first-page":"1105","DOI":"10.18653\/v1\/P16-1105","article-title":"End-to-end relation extraction using LSTMs on sequences and tree structures","volume-title":"ACL","author":"Miwa","year":"2016"},{"key":"2022010319055114200_bib53","doi-asserted-by":"publisher","first-page":"1858","DOI":"10.3115\/v1\/D14-1200","article-title":"Modeling joint entity and relation extraction with table representation","volume-title":"EMNLP","author":"Miwa","year":"2014"},{"key":"2022010319055114200_bib54","first-page":"2270","article-title":"Comparable study of event extraction in newswire and biomedical domains","volume-title":"COLING","author":"Miwa","year":"2014"},{"key":"2022010319055114200_bib55","doi-asserted-by":"publisher","first-page":"355","DOI":"10.1007\/978-94-011-5014-9_12","article-title":"A view of the EM algorithm that justifies incremental, sparse, and other variants","volume-title":"Learning in Graphical Models","author":"Neal","year":"1999"},{"key":"2022010319055114200_bib56","doi-asserted-by":"publisher","first-page":"300","DOI":"10.18653\/v1\/N16-1034","article-title":"Joint event extraction via recurrent neural networks","volume-title":"NAACL","author":"Nguyen","year":"2016"},{"key":"2022010319055114200_bib57","doi-asserted-by":"publisher","first-page":"6851","DOI":"10.1609\/aaai.v33i01.33016851","article-title":"One for all: Neural joint modeling of entities and events","volume-title":"AAAI","author":"Nguyen","year":"2019"},{"issue":"1","key":"2022010319055114200_bib58","doi-asserted-by":"publisher","first-page":"71","DOI":"10.1016\/0004-3702(86)90031-7","article-title":"Probabilistic logic","volume":"28","author":"Nilsson","year":"1986","journal-title":"Artificial Intelligence"},{"key":"2022010319055114200_bib59","doi-asserted-by":"publisher","first-page":"151","DOI":"10.3115\/1699510.1699530","article-title":"A unified model of phrasal and sentential evidence for information extraction","volume-title":"EMNLP","author":"Patwardhan","year":"2009"},{"key":"2022010319055114200_bib60","doi-asserted-by":"publisher","first-page":"27","DOI":"10.3115\/v1\/S14-2004","article-title":"Semeval-2014 task 4: Aspect based sentiment analysis","volume-title":"SemEval","author":"Pontiki","year":"2014"},{"key":"2022010319055114200_bib61","first-page":"813","article-title":"Joint inference for knowledge extraction from biomedical literature","volume-title":"NAACL","author":"Poon","year":"2010"},{"issue":"1","key":"2022010319055114200_bib62","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1162\/coli_a_00034","article-title":"Opinion word expansion and target extraction through double propagation","volume":"37","author":"Qiu","year":"2011","journal-title":"Computational Linguistics"},{"key":"2022010319055114200_bib63","first-page":"5241","article-title":"GMNN: Graph Markov neural networks","volume-title":"ICML","author":"Qu","year":"2019"},{"key":"2022010319055114200_bib64","first-page":"7710","article-title":"Probabilistic logic neural networks for reasoning","volume-title":"NeurIPS","author":"Qu","year":"2019"},{"issue":"1\u20132","key":"2022010319055114200_bib65","doi-asserted-by":"publisher","first-page":"107","DOI":"10.1007\/s10994-006-5833-1","article-title":"Markov logic networks","volume":"62","author":"Richardson","year":"2006","journal-title":"Machine Learning"},{"key":"2022010319055114200_bib66","doi-asserted-by":"publisher","first-page":"41","DOI":"10.3115\/1572340.1572347","article-title":"A Markov Logic approach to bio-molecular event extraction","volume-title":"Proceedings of the BioNLP 2009 Workshop Companion Volume for Shared Task","author":"Riedel","year":"2009"},{"key":"2022010319055114200_bib67","first-page":"3788","article-title":"End-to-end differentiable proving","volume-title":"Advances in Neural Information Processing Systems 30","author":"Rockt\u00e4schel","year":"2017"},{"key":"2022010319055114200_bib68","doi-asserted-by":"publisher","first-page":"1119","DOI":"10.3115\/v1\/N15-1118","article-title":"Injecting logical background knowledge into embeddings for relation extraction","volume-title":"NAACL","author":"Rockt\u00e4schel","year":"2015"},{"key":"2022010319055114200_bib69","first-page":"1","article-title":"A linear programming formulation for global inference in natural language tasks","volume-title":"HLT-NAACL 2004 Workshop: CoNLL-2004","author":"Roth","year":"2004"},{"key":"2022010319055114200_bib70","doi-asserted-by":"crossref","DOI":"10.7551\/mitpress\/7432.003.0022","article-title":"Global inference for entity and relation identification via a linear programming formulation","author":"Roth","year":"2007","journal-title":"Introduction to Statistical Relational Learning"},{"key":"2022010319055114200_bib71","article-title":"Logic tensor networks: Deep learning and logical reasoning from data and knowledge","author":"Serafini","year":"2016","journal-title":"CoRR"},{"key":"2022010319055114200_bib72","doi-asserted-by":"crossref","DOI":"10.1609\/aaai.v32i1.12034","article-title":"Jointly extracting event triggers and arguments by dependency-bridge RNN and tensor-based argument interaction","volume-title":"AAAI","author":"Sha","year":"2018"},{"key":"2022010319055114200_bib73","article-title":"An explicitly relational neural network architecture","author":"Shanahan","year":"2019","journal-title":"CoRR"},{"key":"2022010319055114200_bib74","doi-asserted-by":"publisher","first-page":"2256","DOI":"10.18653\/v1\/D18-1249","article-title":"Extracting entities and relations with joint minimum risk training","volume-title":"EMNLP","author":"Sun","year":"2018"},{"key":"2022010319055114200_bib75","first-page":"7072","article-title":"A hierarchical framework for relation extraction with reinforcement learning","volume-title":"AAAI","author":"Takanobu","year":"2019"},{"issue":"2","key":"2022010319055114200_bib76","doi-asserted-by":"publisher","first-page":"246","DOI":"10.1109\/TNNLS.2016.2603784","article-title":"Deep logic networks: Inserting and extracting knowledge from deep belief networks","volume":"29","author":"Tran","year":"2018","journal-title":"IEEE Transactions on Neural Networks and Learning Systems"},{"key":"2022010319055114200_bib77","doi-asserted-by":"publisher","first-page":"831","DOI":"10.3115\/v1\/D14-1090","article-title":"Relieving the computational bottleneck: Joint inference for event extraction with high-dimensional features","volume-title":"EMNLP","author":"Venugopal","year":"2014"},{"key":"2022010319055114200_bib78","doi-asserted-by":"publisher","first-page":"5784","DOI":"10.18653\/v1\/D19-1585","article-title":"Entity, relation, and event extraction with contextualized span representations","volume-title":"Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)","author":"Wadden","year":"2019"},{"key":"2022010319055114200_bib79","doi-asserted-by":"publisher","first-page":"1891","DOI":"10.18653\/v1\/D18-1215","article-title":"Deep probabilistic logic: A unifying framework for indirect supervision","volume-title":"EMNLP","author":"Wang","year":"2018"},{"key":"2022010319055114200_bib80","doi-asserted-by":"publisher","first-page":"1706","DOI":"10.18653\/v1\/2020.emnlp-main.133","article-title":"Two are better than one: Joint entity and relation extraction with table-sequence encoders","volume-title":"Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Wang","year":"2020"},{"key":"2022010319055114200_bib81","first-page":"6545","article-title":"SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solver","volume-title":"ICML","author":"Wang","year":"2019"},{"key":"2022010319055114200_bib82","doi-asserted-by":"publisher","first-page":"4461","DOI":"10.24963\/ijcai.2018\/620","article-title":"Joint extraction of entities and relations based on a novel graph scheme","volume-title":"IJCAI","author":"Wang","year":"2018"},{"key":"2022010319055114200_bib83","doi-asserted-by":"publisher","first-page":"9225","DOI":"10.1609\/aaai.v34i05.6460","article-title":"Integrating deep learning with logic fusion for information extraction","volume-title":"AAAI","author":"Wang","year":"2020"},{"key":"2022010319055114200_bib84","doi-asserted-by":"publisher","first-page":"616","DOI":"10.18653\/v1\/D16-1059","article-title":"Recursive neural conditional random fields for aspect-based sentiment analysis","volume-title":"EMNLP","author":"Wang","year":"2016"},{"key":"2022010319055114200_bib85","first-page":"3316","article-title":"Coupled multi-layer tensor network for co-extraction of aspect and opinion terms","volume-title":"AAAI","author":"Wang","year":"2017"},{"key":"2022010319055114200_bib86","doi-asserted-by":"publisher","first-page":"592","DOI":"10.18653\/v1\/P18-2094","article-title":"Double embeddings and CNN-based sequence labeling for aspect extraction","volume-title":"ACL","author":"Xu","year":"2018"},{"key":"2022010319055114200_bib87","first-page":"5502","article-title":"A semantic loss function for deep learning with symbolic knowledge","volume-title":"ICML","author":"Xu","year":"2018"},{"key":"2022010319055114200_bib88","first-page":"1640","article-title":"Joint inference for fine-grained opinion extraction","volume-title":"ACL","author":"Yang","year":"2013"},{"key":"2022010319055114200_bib89","first-page":"289","article-title":"Joint extraction of events and entities within a document context","volume-title":"NAACL","author":"Yang","year":"2016"},{"key":"2022010319055114200_bib90","article-title":"Learn to explain efficiently via neural logic inductive learning","author":"Yang","year":"2020","journal-title":"ICLR"},{"key":"2022010319055114200_bib91","first-page":"2979","article-title":"Unsupervised word and dependency path embeddings for aspect term extraction","volume-title":"IJCAI","author":"Yin","year":"2016"},{"issue":"1","key":"2022010319055114200_bib92","doi-asserted-by":"publisher","first-page":"168","DOI":"10.1109\/TASLP.2018.2875170","article-title":"Global inference for aspect and opinion terms co-extraction based on multi-task neural networks","volume":"27","author":"Yu","year":"2019","journal-title":"IEEE\/ACM Transactions on Audio, Speech, and Language Processing"},{"key":"2022010319055114200_bib93","first-page":"1399","article-title":"Jointly identifying entities and extracting relations in encyclopedia text via a graphical model approach","volume-title":"COLING","author":"Yu","year":"2010"},{"key":"2022010319055114200_bib94","doi-asserted-by":"publisher","first-page":"506","DOI":"10.18653\/v1\/P18-1047","article-title":"Extracting relational facts by an end-to-end neural model with copy mechanism","volume-title":"ACL","author":"Zeng","year":"2018"},{"key":"2022010319055114200_bib95","doi-asserted-by":"publisher","first-page":"1730","DOI":"10.18653\/v1\/D17-1182","article-title":"End-to-end neural relation extraction with global optimization","volume-title":"EMNLP","author":"Zhang","year":"2017"},{"issue":"2","key":"2022010319055114200_bib96","doi-asserted-by":"publisher","first-page":"99","DOI":"10.1162\/dint_a_00014","article-title":"Joint entity and event extraction with generative adversarial imitation learning","volume":"1","author":"Zhang","year":"2019","journal-title":"Data Intelligence"},{"key":"2022010319055114200_bib97","first-page":"1227","article-title":"Joint extraction of entities and relations based on a novel tagging scheme","volume-title":"ACL","author":"Zheng","year":"2017"}],"container-title":["Computational Linguistics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/coli\/article-pdf\/47\/4\/775\/1979406\/coli_a_00415.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/coli\/article-pdf\/47\/4\/775\/1979406\/coli_a_00415.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,6]],"date-time":"2023-11-06T23:21:01Z","timestamp":1699312861000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/coli\/article\/47\/4\/775\/106773\/Variational-Deep-Logic-Network-for-Joint-Inference"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,12]]},"references-count":97,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2021,12,23]]},"published-print":{"date-parts":[[2021,12,23]]}},"URL":"https:\/\/doi.org\/10.1162\/coli_a_00415","relation":{},"ISSN":["0891-2017","1530-9312"],"issn-type":[{"value":"0891-2017","type":"print"},{"value":"1530-9312","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2021,12]]},"published":{"date-parts":[[2021,12]]}}}