{"id":"https://openalex.org/W3094353525","doi":"https://doi.org/10.1162/coli_r_00389","title":"<i>Deep Learning Approaches to Text Production</i>. By Shashi Narayan and Claire Gardent (University of Edinburgh; CNRS/LORIA, Nancy). Morgan &amp; Claypool (Synthesis Lectures on Human Language Technologies, edited by Graeme Hirst, volume 43), 2020, xxvi+175 pp; paperback, ISBN 978-1-68173-758-4, $79.95; ebook, ISBN 978-1-68173-759-1, $63.96; hardcover, ISBN 978-1-68173-760-7, $99.95; doi:10.2200/S00979ED1V01Y201912HLT044","display_name":"<i>Deep Learning Approaches to Text Production</i>. By Shashi Narayan and Claire Gardent (University of Edinburgh; CNRS/LORIA, Nancy). Morgan &amp; Claypool (Synthesis Lectures on Human Language Technologies, edited by Graeme Hirst, volume 43), 2020, xxvi+175 pp; paperback, ISBN 978-1-68173-758-4, $79.95; ebook, ISBN 978-1-68173-759-1, $63.96; hardcover, ISBN 978-1-68173-760-7, $99.95; doi:10.2200/S00979ED1V01Y201912HLT044","publication_year":2020,"publication_date":"2020-10-20","ids":{"openalex":"https://openalex.org/W3094353525","doi":"https://doi.org/10.1162/coli_r_00389","mag":"3094353525"},"language":"en","primary_location":{"id":"doi:10.1162/coli_r_00389","is_oa":true,"landing_page_url":"https://doi.org/10.1162/coli_r_00389","pdf_url":"https://direct.mit.edu/coli/article-pdf/46/4/899/1888270/coli_r_00389.pdf","source":{"id":"https://openalex.org/S155526855","display_name":"Computational Linguistics","issn_l":"0891-2017","issn":["0891-2017","1530-9312"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"journal"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computational Linguistics","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://direct.mit.edu/coli/article-pdf/46/4/899/1888270/coli_r_00389.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100333729","display_name":"Yue Zhang","orcid":"https://orcid.org/0000-0002-5214-2268"},"institutions":[{"id":"https://openalex.org/I3133055985","display_name":"Westlake University","ror":"https://ror.org/05hfa4n20","country_code":"CN","type":"education","lineage":["https://openalex.org/I3133055985"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Yue Zhang","raw_affiliation_strings":["Westlake University, Westlake Institute for Advanced Study"],"affiliations":[{"raw_affiliation_string":"Westlake University, Westlake Institute for Advanced Study","institution_ids":["https://openalex.org/I3133055985"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5100333729"],"corresponding_institution_ids":["https://openalex.org/I3133055985"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.13471975,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"46","issue":"4","first_page":"899","last_page":"903"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.9914000034332275,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.4755922555923462},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.46325549483299255},{"id":"https://openalex.org/keywords/cognitive-science","display_name":"Cognitive science","score":0.3479539453983307},{"id":"https://openalex.org/keywords/sociology","display_name":"Sociology","score":0.34593167901039124},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.33668452501296997},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.27653348445892334}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4755922555923462},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46325549483299255},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.3479539453983307},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"score":0.34593167901039124},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.33668452501296997},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.27653348445892334}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1162/coli_r_00389","is_oa":true,"landing_page_url":"https://doi.org/10.1162/coli_r_00389","pdf_url":"https://direct.mit.edu/coli/article-pdf/46/4/899/1888270/coli_r_00389.pdf","source":{"id":"https://openalex.org/S155526855","display_name":"Computational Linguistics","issn_l":"0891-2017","issn":["0891-2017","1530-9312"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"journal"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computational Linguistics","raw_type":"journal-article"},{"id":"pmh:oai:doaj.org/article:aa629299f44c4e799d92a90b7c6b9b88","is_oa":true,"landing_page_url":"https://doaj.org/article/aa629299f44c4e799d92a90b7c6b9b88","pdf_url":null,"source":{"id":"https://openalex.org/S112646816","display_name":"SHILAP Revista de lepidopterolog\u00eda","issn_l":"0300-5267","issn":["0300-5267","2340-4078"],"is_oa":true,"is_in_doaj":true,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Computational Linguistics, Vol 46, Iss 4, Pp 899-903 (2021)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.1162/coli_r_00389","is_oa":true,"landing_page_url":"https://doi.org/10.1162/coli_r_00389","pdf_url":"https://direct.mit.edu/coli/article-pdf/46/4/899/1888270/coli_r_00389.pdf","source":{"id":"https://openalex.org/S155526855","display_name":"Computational Linguistics","issn_l":"0891-2017","issn":["0891-2017","1530-9312"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"journal"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computational Linguistics","raw_type":"journal-article"},"sustainable_development_goals":[{"display_name":"Quality Education","score":0.7099999785423279,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3094353525.pdf","grobid_xml":"https://content.openalex.org/works/W3094353525.grobid-xml"},"referenced_works_count":24,"referenced_works":["https://openalex.org/W155168364","https://openalex.org/W1832693441","https://openalex.org/W2064675550","https://openalex.org/W2157331557","https://openalex.org/W2250539671","https://openalex.org/W2507756961","https://openalex.org/W2557283755","https://openalex.org/W2559655401","https://openalex.org/W2604272474","https://openalex.org/W2604799547","https://openalex.org/W2611972942","https://openalex.org/W2612690371","https://openalex.org/W2626778328","https://openalex.org/W2752446824","https://openalex.org/W2884814595","https://openalex.org/W2896457183","https://openalex.org/W2949888546","https://openalex.org/W2950133940","https://openalex.org/W2962739339","https://openalex.org/W2963260202","https://openalex.org/W2963374482","https://openalex.org/W2963653811","https://openalex.org/W2964308564","https://openalex.org/W3124386704"],"related_works":["https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W2382290278","https://openalex.org/W2350741829","https://openalex.org/W2530322880","https://openalex.org/W1596801655","https://openalex.org/W2359140296"],"abstract_inverted_index":{"Text":[0],"production":[1,1238],"(Reiter":[2],"and":[3,7,72,80,112,129,197,213,227,253,270,282,305,319,352,369,373,383,400,411,434,447,494,527,567,602,643,648,651,669,673,676,690,706,712,735,740,764,787,805,840,863,866,869,880,899,905,908,967,985,994,1006,1030,1037,1040,1067,1099,1126,1184,1254,1264,1272,1293,1307,1328,1333,1346,1371,1374,1378,1387,1426,1438,1457],"Dale":[4],"2000;":[5],"Gatt":[6],"Krahmer":[8],"2018)":[9,1066],"is":[10,20,154,237,276,367,518,530,541,760,1226,1241],"also":[11,658,716,753,923,954,1294],"referred":[12],"to":[13,60,88,100,149,182,285,316,323,396,472,503,547,595,784,791,819,887,891,931,937,962,1314,1418],"as":[14,36,38,77,136,246,298,391,432,488,524,599,745,796,947,1095,1309,1348,1392,1413],"natural":[15,24,32,39],"language":[16,25,33,40],"generation":[17,30,340,347,353,398],"(NLG).":[18],"It":[19,415,475,584,621,657,883,970,1012,1240,1300,1410,1433],"a":[21,61,119,137,145,151,160,364,484,525,534,573,702,707,730,850,928,948,973,1015,1097,1101,1163,1178,1281,1310,1398,1414],"subtask":[22],"of":[23,31,54,63,194,203,223,291,331,361,464,623,634,749,812,825,852,975,1003,1043,1051,1057,1111,1141,1165,1181,1220,1283,1292,1317,1358,1401,1485],"processing":[26],"focusing":[27],"on":[28,92,265,502,577,800,1118,1130,1249],"the":[29,52,114,192,201,209,214,266,268,279,286,289,292,311,329,359,362,370,381,386,403,412,438,452,462,465,505,552,607,611,631,698,750,768,771,775,792,797,810,822,834,837,940,978,999,1052,1055,1120,1127,1132,1152,1206,1214,1218,1230,1315,1318,1322,1325,1330,1341,1356,1402,1419,1422,1474,1478,1482,1486],"text.":[34],"Although":[35,1440],"important":[37,481],"understanding":[41,318,461],"for":[42,73,426,460,483,508,532,560,605,618,721,782,815,832,841,858,956,982,1020,1032,1045,1089,1155,1223,1269,1305,1421],"communication,":[43],"NLG":[44,86,98,218,388,942,1084],"had":[45,423,592,1459],"received":[46],"relatively":[47],"less":[48],"research":[49,64,1424,1487],"attention.":[50],"Recently,":[51],"rise":[53],"deep":[55,635,801],"learning":[56,84,802,1208],"techniques":[57,166,296,559,720,831,952,1088,1327,1405],"has":[58,132],"led":[59],"surge":[62],"interest":[65],"in":[66,70,232,451,724,762,939,996,1008,1075,1082,1104,1151,1187,1200,1235,1243,1276,1406,1462,1477],"text":[67,78,152,188,205,332,355,397,473,619,966,1098,1237],"production,":[68,206,333],"both":[69,195,208,1034],"general":[71,1179],"specific":[74],"applications":[75,1186],"such":[76,245,297,390,431,487,598,795,1094,1494],"summarization":[79,433,1225],"dialogue":[81,405,992],"systems.":[82],"Deep":[83],"allows":[85],"models":[87,181,1019,1159,1167],"be":[89,582,780,1302],"constructed":[90],"based":[91],"neural":[93,120,146,161,180,204,216,242,248,255,280,613,639,722,736,941,1201,1236,1353,1408],"representations,":[94],"thereby":[95],"enabling":[96],"end-to-end":[97],"systems":[99],"replace":[101],"traditional":[102,283,485],"pipeline":[103],"approaches,":[104,737],"which":[105,140,150,509,580,766,894,934,1023,1247],"frees":[106],"us":[107],"from":[108,185,341,348,354,742],"tedious":[109],"engineering":[110,744],"efforts":[111],"improves":[113],"output":[115],"quality.":[116],"In":[117,1339],"particular,":[118,1340],"encoder-decoder":[121,235,616,699,856,980,1121],"structure":[122,981],"(Cho":[123,258,653],"et":[124,168,174,259,302,408,654,666,682,687,911,1064,1070,1384],"al.":[125,169,175,260,303,409,655,667,683,688,912,1065,1071,1385],"2014;":[126],"Sutskever,":[127],"Vinyals,":[128],"Le":[130],"2014)":[131,252],"been":[133,418,424,587,593,1452,1460],"widely":[134],"used":[135,523,546,1234],"basic":[138,241,615,632,1430],"framework,":[139],"computes":[141],"input":[142,337],"representations":[143,343],"using":[144,159,701,855,945,977,1028,1060],"encoder,":[147,267],"according":[148],"sequence":[153],"generated":[155],"token":[156,158],"by":[157],"decoder.":[162,710],"Very":[163],"recently,":[164],"pre-training":[165,306,1388,1458],"(Broscheit":[167],"2010;":[170],"Radford":[171],"2018;":[172],"Devlin":[173],"2019)":[176],"have":[177,417,586,1451],"further":[178,190,325,816],"allowed":[179],"collect":[183],"knowledge":[184,1379],"large":[186,535],"raw":[187],"data,":[189],"improving":[191,833],"quality":[193],"encoding":[196,838,983],"decoding.This":[198],"book":[199,365,799,1396,1417],"introduces":[200,328,630,697],"fundamentals":[202],"discussing":[207,334,637,754,1017,1329],"mostly":[210],"investigated":[211],"tasks":[212,219,389,430,1189],"foundational":[215,612,1407],"methods.":[217,1361],"with":[220,240,477,849,964,972,1123,1135,1190,1352,1429,1465],"different":[221,1191],"types":[222,1110],"inputs":[224,1370],"are":[225,230,263,307,314,321,376,442,448,497,513,522,545,715,922,935,953,1113,1160,1168,1267,1274,1389,1473],"introduced,":[226],"benchmark":[228],"datasets":[229,1233],"discussed":[231,449,1074,1391],"detail.":[233],"The":[234,515,538,627,694,726,925,1048,1115,1138,1145,1171,1194,1258],"architecture":[236,1122],"introduced":[238],"together":[239],"network":[243,249,256,1026],"components":[244,482],"convolutional":[247],"(CNN)":[250],"(Kim":[251],"recurrent":[254],"(RNN)":[257],"2014).":[261,656],"Elaborations":[262],"given":[264,315,425,594,1268],"decoder,":[269],"task-specific":[271,843,1091,1124,1136],"optimization":[272],"techniques.":[273],"A":[274],"contrast":[275],"made":[277],"between":[278,732],"solution":[281],"solutions":[284,890],"task.":[287],"Toward":[288],"end":[290,360],"book,":[293,312,1323],"more":[294,421,590,1080,1453,1463],"recent":[295,1337,1393],"self-attention":[299,957],"networks":[300,723],"(Vaswani":[301,1383],"2017)":[304,1386],"briefly":[308,377,468,659,1390],"discussed.":[309,378,498,514,924],"Throughout":[310],"figures":[313,384,763],"facilitate":[317],"references":[320,591],"provided":[322],"enable":[324],"reading.Chapter":[326],"1":[327],"task":[330,406],"three":[335,624,885,1142,1244,1365],"typical":[336,387,429,1018],"settings,":[338],"namely,":[339,876,1368],"meaning":[342,393],"(MR;":[344],"i.e.,":[345],"realization),":[346],"data":[349,1261],"(i.e.,":[350,356,663,679,860],"data-to-text),":[351],"text-to-text).":[357],"At":[358],"chapter,":[363],"outline":[366],"given,":[368],"scope,":[371],"coverage,":[372,879],"notation":[374],"convention":[375],"I":[379,454,1204,1289,1296,1441,1446],"enjoyed":[380,1442],"examples":[382,422,446,1273],"demonstrating":[385,1217],"abstract":[392],"representation":[394],"(AMR)":[395],"(May":[399],"Priyadarshi":[401],"2017),":[402],"E2E":[404],"(Li":[407],"2018),":[410],"data-to-text":[413,478,596,867,1250],"examples.":[414],"would":[416,585,1450],"useful":[419,459,588,955,1304],"if":[420,589,1455],"some":[427,1295],"other":[428,776,1166,1265],"sentence":[435,561,563,565],"compression,":[436,564],"despite":[437],"fact":[439],"that":[440,1288,1448],"they":[441,788,1363,1472],"intuitively":[443],"understandable":[444],"without":[445],"later":[450],"book.":[453,827],"find":[455,1205],"Section":[456,813],"1.3":[457],"particularly":[458,1210],"scope":[463],"book.Chapter":[466,1319],"2":[467],"summarizes":[469,1321],"pre-neural":[470,733],"approaches":[471,512,601,604,734,921],"production.":[474],"begins":[476,848,971],"generation,":[479,507,556,1251,1253,1256],"where":[480,520,543],"pipeline,":[486],"content":[489,1149],"selection,":[490],"document":[491,568],"planning,":[492],"lexicalization,":[493],"surface":[495],"realization,":[496],"Then":[499],"it":[500,778,1449],"moves":[501],"discuss":[504,829],"MR-to-text":[506,1252],"two":[510,746,1109,1349],"major":[511,558,747,830,889,1350],"first":[516,628,1116,1146],"approach":[517,540],"grammar-centric,":[519],"rules":[521],"basis":[526],"much":[528,1079,1313],"care":[529],"taken":[531],"pruning":[533],"search":[536],"space.":[537],"second":[539,695,1128,1172],"statistical,":[542],"features":[544,1434],"score":[548],"candidate":[549],"outputs.":[550],"Finally,":[551,1381],"chapter":[553,571,759,926,1053,1139,1279],"discusses":[554,610,660,960,1054,1087,1148,1174,1197],"text-to-text":[555,578,859,1255],"introducing":[557,888],"simplification,":[562],"paraphrasing,":[566],"summarization.":[569,1156],"This":[570,758,1278],"presents":[572,1397],"rich":[574,761,1435],"literature":[575,1467],"review":[576,974,1400],"methods,":[579,597],"can":[581,779,1301,1411],"helpful.":[583],"modular":[600],"integrated":[603],"implementing":[606],"pipeline.Chapter":[608],"3":[609],"model\u2014a":[614],"framework":[617,700],"generation.":[620],"consists":[622,1140],"main":[625,874,1000,1143,1245,1326,1366],"sections.":[626,1144],"section":[629,696,728,1016,1050,1077,1147,1173,1196,1209],"elements":[633],"learning,":[636,1176,1377],"feed-forward":[638],"networks,":[640],"CNNs,":[641],"RNNs,":[642],"their":[644,755],"variants":[645],"LSTM":[646],"(Hochreiter":[647],"Schmidhuber":[649],"1997)":[650],"GRU":[652],"word":[661],"embeddings":[662,678],"word2vec":[664],"[Mikolov":[665],"2013]":[668],"GloVe":[670],"[Pennington,":[671],"Socher,":[672],"Manning":[674],"2014])":[675],"contextualized":[677],"ELMo":[680],"[Peters":[681],"2018],":[684],"BERT":[685],"[Devlin":[686],"2019],":[689],"GPT":[691],"[Radford":[692],"2018]).":[693],"bidirectional":[703],"RNN":[704,709,946],"encoder":[705,1153],"simple":[708],"Training":[711],"decoding":[713,835],"issues":[714,1332,1351],"discussed,":[717],"including":[718,1285],"training":[719,1133],"general.":[725],"final":[727,1049],"makes":[729],"comparison":[731],"highlighting":[738,1355],"robustness":[739],"freedom":[741],"feature":[743],"advantages":[748],"latter,":[751],"while":[752,1162],"potential":[756],"limitations.":[757],"references,":[765],"helps":[767],"reader":[769],"understand":[770],"big":[772],"picture.":[773],"On":[774],"hand,":[777],"difficult":[781],"beginners":[783],"fully":[785],"absorb,":[786],"should":[789],"refer":[790],"reference":[793,1436],"materials":[794,1437],"Goodfellow":[798],"(Goodfellow,":[803],"Bengio,":[804],"Courville":[806],"2016)":[807,913],"cited":[808],"at":[809],"beginning":[811],"3.1":[814],"reading.Chapters":[817],"4":[818,847],"6":[820,1086],"form":[821],"central":[823],"part":[824],"this":[826,1076,1107,1395],"They":[828],"module,":[836,839],"integrating":[842,1090],"objectives,":[844],"respectively.":[845,1257],"Chapter":[846],"survey":[851],"seminal":[853],"work":[854],"modeling":[857,1033,1042,1056,1199],"machine":[861],"translation":[862],"summarization),":[864],"MR-to-text,":[865],"tasks,":[868],"then":[870,1013],"lays":[871],"out":[872],"four":[873],"issues,":[875,893],"accuracy,":[877],"repetitions,":[878],"rare/unknown":[881],"words.":[882],"devotes":[884],"sections":[886],"these":[892,932,951],"include":[895,1024],"attention":[896,1081],"(Bahdanau,":[897],"Cho,":[898],"Bengio":[900],"2015),":[901,907],"copy":[902],"(Vinyals,":[903],"Fortunato,":[904],"Jaitly":[906],"coverage":[909],"(Tu":[910],"mechanisms.":[914],"For":[915,1212],"each":[916,1270],"method,":[917],"similar":[918],"or":[919],"alternative":[920],"gives":[927],"concise":[929],"introduction":[930],"techniques,":[933],"essential":[936],"know":[938],"literature.":[943,1480],"Though":[944],"base":[949],"model,":[950],"networks.Chapter":[958],"5":[959],"how":[961],"deal":[963],"long":[965,1369],"graph-structured":[968],"data.":[969],"methods":[976,1112,1476],"standard":[979],"documents":[984],"linearized":[986],"graphs":[987],"(e.g.,":[988],"AMR,":[989],"RDF":[990],"triples,":[991],"moves,":[993],"Infoboxes":[995],"Wikipedia),":[997],"showing":[998],"limitation:":[1001],"lack":[1002],"structural":[1004],"information":[1005],"weakness":[1007],"capturing":[1009],"long-range":[1010],"dependencies.":[1011],"spends":[1014],"long-text":[1021],"structures,":[1022,1039],"hierarchical":[1025],"structures":[1027,1036,1059],"RNNs":[1029],"CNNs":[1031],"word-sentence":[1035],"sentence-document":[1038],"collaborative":[1041],"paragraphs":[1044],"representing":[1046],"documents.":[1047],"graph":[1058,1061],"LSTMs":[1062],"(Song":[1063],"GCNs":[1068],"(Bastings":[1069],"2017).":[1072],"Techniques":[1073],"receive":[1078],"current":[1083,1479],"research.Chapter":[1085],"communication":[1092],"goals":[1093],"summarizing":[1096],"generating":[1100],"user-specific":[1102],"response":[1103],"dialogue.":[1105],"To":[1106],"end,":[1108],"introduced.":[1114],"focuses":[1117,1129],"augmenting":[1119,1131],"features,":[1125],"objective":[1134],"metrics.":[1137],"selection":[1150],"module":[1154],"Several":[1157],"representative":[1158],"detailed":[1161],"range":[1164,1282],"surveyed":[1169],"briefly.":[1170],"reinforcement":[1175,1207],"describing":[1177],"algorithm":[1180],"policy":[1182],"gradient":[1183],"its":[1185,1444],"many":[1188],"reward":[1192],"functions.":[1193],"third":[1195],"user":[1198],"conversational":[1202],"models.":[1203],"informative.":[1211],"example,":[1213],"case":[1215],"study":[1216],"disadvantage":[1219],"cross-entropy":[1221],"loss":[1222],"extractive":[1224],"insightful.Chapter":[1227],"7":[1228],"describes":[1229],"most":[1231,1286,1403],"prominent":[1232,1404],"research.":[1239],"organized":[1242],"sections,":[1246],"focus":[1248],"origin,":[1259],"size,":[1260],"source,":[1262],"format,":[1263],"characteristics":[1266],"dataset,":[1271],"shown":[1275],"figures.":[1277,1439],"covers":[1280],"datasets,":[1284],"benchmarks":[1287],"am":[1290,1297],"aware":[1291],"unfamiliar":[1298],"with.":[1299],"highly":[1303],"researchers":[1306],"students":[1308],"reference,":[1311],"adding":[1312],"value":[1316],"8":[1320],"reviewing":[1324],"remaining":[1331],"challenges,":[1334,1367],"before":[1335],"mentioning":[1336],"trends.":[1338],"authors":[1342],"identify":[1343],"semantic":[1344],"adequacy":[1345],"explainability":[1347],"NLG,":[1354],"limitation":[1357],"existing":[1359],"evaluation":[1360],"Additionally,":[1362],"raise":[1364],"outputs,":[1372],"cross-domain":[1373],"cross-lingual":[1375],"transfer":[1376],"integration.":[1380],"Transformer":[1382,1456],"trends.Overall,":[1394],"succinct":[1399],"NLG.":[1409],"serve":[1412],"great":[1415],"introductory":[1416],"field":[1420],"NLP":[1423,1427],"community":[1425],"engineers":[1428],"relevant":[1431,1466],"background.":[1432],"reading":[1443],"content,":[1445],"feel":[1447],"valuable":[1454],"elaborated":[1461],"detail,":[1464],"surveys":[1468],"being":[1469],"included,":[1470],"since":[1471],"dominant":[1475],"Given":[1481],"fast-moving":[1483],"pace":[1484],"field,":[1488],"maybe":[1489],"subsequent":[1490],"editions":[1491],"will":[1492],"meet":[1493],"expectations.":[1495]},"counts_by_year":[],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2020-10-29T00:00:00"}