{"id":"https://openalex.org/W3110014757","doi":"https://doi.org/10.1109/tpami.2020.3042192","title":"Auto-encoding and Distilling Scene Graphs for Image Captioning","display_name":"Auto-encoding and Distilling Scene Graphs for Image Captioning","publication_year":2020,"publication_date":"2020-01-01","ids":{"openalex":"https://openalex.org/W3110014757","doi":"https://doi.org/10.1109/tpami.2020.3042192","mag":"3110014757","pmid":"https://pubmed.ncbi.nlm.nih.gov/33270557"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2020.3042192","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2020.3042192","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","pubmed"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5064752243","display_name":"Xu Yang","orcid":"https://orcid.org/0000-0002-8276-2679"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":true,"raw_author_name":"Xu Yang","raw_affiliation_strings":["Multimedia and Interactive Computing Lab, School of Computer Science and Engineering, Nanyang Technological University, Singapore, Singapore"],"affiliations":[{"raw_affiliation_string":"Multimedia and Interactive Computing Lab, School of Computer Science and Engineering, Nanyang Technological University, Singapore, Singapore","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042324027","display_name":"Hanwang Zhang","orcid":"https://orcid.org/0000-0001-7374-8739"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Hanwang Zhang","raw_affiliation_strings":["Nanyang Technological University, Singapore, Singapore"],"affiliations":[{"raw_affiliation_string":"Nanyang Technological University, Singapore, Singapore","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100635804","display_name":"Jianfei Cai","orcid":"https://orcid.org/0000-0002-9444-3763"},"institutions":[{"id":"https://openalex.org/I56590836","display_name":"Monash University","ror":"https://ror.org/02bfwt286","country_code":"AU","type":"education","lineage":["https://openalex.org/I56590836"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Jianfei Cai","raw_affiliation_strings":["Data Science & AI Department, Faculty of IT, Monash University, Clayton, VIC, Australia"],"affiliations":[{"raw_affiliation_string":"Data Science & AI Department, Faculty of IT, Monash University, Clayton, VIC, Australia","institution_ids":["https://openalex.org/I56590836"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5064752243"],"corresponding_institution_ids":["https://openalex.org/I172675005"],"apc_list":null,"apc_paid":null,"fwci":3.8272,"has_fulltext":false,"cited_by_count":68,"citation_normalized_percentile":{"value":0.94815208,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":100},"biblio":{"volume":"44","issue":"5","first_page":"1","last_page":"1"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10028","display_name":"Topic Modeling","score":0.9861000180244446,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9854000210762024,"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":0.8256272077560425},{"id":"https://openalex.org/keywords/scene-graph","display_name":"Scene graph","score":0.8102469444274902},{"id":"https://openalex.org/keywords/closed-captioning","display_name":"Closed captioning","score":0.7343960404396057},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.6381768584251404},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5611586570739746},{"id":"https://openalex.org/keywords/sentence","display_name":"Sentence","score":0.5604568123817444},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.5427477955818176},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5338958501815796},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.5151344537734985},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.5079420208930969},{"id":"https://openalex.org/keywords/inductive-bias","display_name":"Inductive bias","score":0.4765104353427887},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.4699295461177826},{"id":"https://openalex.org/keywords/factor-graph","display_name":"Factor graph","score":0.4519970417022705},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.43439194560050964},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.41810745000839233},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.4149931073188782},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.4061478078365326},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.32377272844314575},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.21458613872528076},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.11188071966171265},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.10013782978057861}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8256272077560425},{"id":"https://openalex.org/C179372163","wikidata":"https://www.wikidata.org/wiki/Q1406181","display_name":"Scene graph","level":3,"score":0.8102469444274902},{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.7343960404396057},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.6381768584251404},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5611586570739746},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.5604568123817444},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5427477955818176},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5338958501815796},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.5151344537734985},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.5079420208930969},{"id":"https://openalex.org/C197352929","wikidata":"https://www.wikidata.org/wiki/Q1074074","display_name":"Inductive bias","level":4,"score":0.4765104353427887},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.4699295461177826},{"id":"https://openalex.org/C159246509","wikidata":"https://www.wikidata.org/wiki/Q5428725","display_name":"Factor graph","level":3,"score":0.4519970417022705},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.43439194560050964},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.41810745000839233},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.4149931073188782},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4061478078365326},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.32377272844314575},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.21458613872528076},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.11188071966171265},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.10013782978057861},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.0},{"id":"https://openalex.org/C28006648","wikidata":"https://www.wikidata.org/wiki/Q6934509","display_name":"Multi-task learning","level":3,"score":0.0},{"id":"https://openalex.org/C205711294","wikidata":"https://www.wikidata.org/wiki/Q176953","display_name":"Rendering (computer graphics)","level":2,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[{"descriptor_ui":"D000465","descriptor_name":"Algorithms","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D000465","descriptor_name":"Algorithms","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D000465","descriptor_name":"Algorithms","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D006801","descriptor_name":"Humans","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D006801","descriptor_name":"Humans","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D006801","descriptor_name":"Humans","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D007858","descriptor_name":"Learning","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D007858","descriptor_name":"Learning","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D007858","descriptor_name":"Learning","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D019985","descriptor_name":"Benchmarking","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D019985","descriptor_name":"Benchmarking","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D019985","descriptor_name":"Benchmarking","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false}],"locations_count":2,"locations":[{"id":"doi:10.1109/tpami.2020.3042192","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2020.3042192","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},{"id":"pmid:33270557","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/33270557","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE transactions on pattern analysis and machine intelligence","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.4699999988079071,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":106,"referenced_works":["https://openalex.org/W637153065","https://openalex.org/W1522301498","https://openalex.org/W1662382123","https://openalex.org/W1821462560","https://openalex.org/W1861492603","https://openalex.org/W1895577753","https://openalex.org/W1905882502","https://openalex.org/W1931639407","https://openalex.org/W1956340063","https://openalex.org/W1969616664","https://openalex.org/W2077069816","https://openalex.org/W2097606805","https://openalex.org/W2101105183","https://openalex.org/W2105482032","https://openalex.org/W2139906443","https://openalex.org/W2154652894","https://openalex.org/W2176263492","https://openalex.org/W2194775991","https://openalex.org/W2250378130","https://openalex.org/W2277195237","https://openalex.org/W2302086703","https://openalex.org/W2326588846","https://openalex.org/W2473930607","https://openalex.org/W2506483933","https://openalex.org/W2552161745","https://openalex.org/W2558834163","https://openalex.org/W2561238782","https://openalex.org/W2561715562","https://openalex.org/W2570343428","https://openalex.org/W2575842049","https://openalex.org/W2579549467","https://openalex.org/W2591644541","https://openalex.org/W2600702321","https://openalex.org/W2745461083","https://openalex.org/W2754927243","https://openalex.org/W2795151422","https://openalex.org/W2798441115","https://openalex.org/W2805516822","https://openalex.org/W2885013662","https://openalex.org/W2886970679","https://openalex.org/W2887029921","https://openalex.org/W2887585070","https://openalex.org/W2890531016","https://openalex.org/W2896457183","https://openalex.org/W2938603906","https://openalex.org/W2950898568","https://openalex.org/W2962779575","https://openalex.org/W2963037989","https://openalex.org/W2963048642","https://openalex.org/W2963062932","https://openalex.org/W2963084599","https://openalex.org/W2963101956","https://openalex.org/W2963150697","https://openalex.org/W2963184176","https://openalex.org/W2963305465","https://openalex.org/W2963448089","https://openalex.org/W2963448850","https://openalex.org/W2963536419","https://openalex.org/W2963656855","https://openalex.org/W2963743213","https://openalex.org/W2963762755","https://openalex.org/W2963921921","https://openalex.org/W2963938081","https://openalex.org/W2964082701","https://openalex.org/W2964189064","https://openalex.org/W2964203871","https://openalex.org/W2965697393","https://openalex.org/W2965833116","https://openalex.org/W2969679616","https://openalex.org/W2984121207","https://openalex.org/W2987123286","https://openalex.org/W2990069284","https://openalex.org/W2990307191","https://openalex.org/W2992478697","https://openalex.org/W2999219213","https://openalex.org/W3105136412","https://openalex.org/W4297749157","https://openalex.org/W6600334730","https://openalex.org/W6620673361","https://openalex.org/W6620707391","https://openalex.org/W6621543089","https://openalex.org/W6630875275","https://openalex.org/W6631190155","https://openalex.org/W6637178625","https://openalex.org/W6637306801","https://openalex.org/W6638318767","https://openalex.org/W6638523607","https://openalex.org/W6678262379","https://openalex.org/W6679436768","https://openalex.org/W6679844565","https://openalex.org/W6682086108","https://openalex.org/W6682132143","https://openalex.org/W6682631176","https://openalex.org/W6685322675","https://openalex.org/W6690815549","https://openalex.org/W6697449767","https://openalex.org/W6713582119","https://openalex.org/W6717697761","https://openalex.org/W6719270105","https://openalex.org/W6730179637","https://openalex.org/W6748163181","https://openalex.org/W6748270630","https://openalex.org/W6748581942","https://openalex.org/W6751796012","https://openalex.org/W6754929296","https://openalex.org/W6755207826"],"related_works":["https://openalex.org/W4210416330","https://openalex.org/W3088136942","https://openalex.org/W2949362007","https://openalex.org/W2775506363","https://openalex.org/W4287661223","https://openalex.org/W2902243109","https://openalex.org/W2965288565","https://openalex.org/W2995273010","https://openalex.org/W2998512808","https://openalex.org/W4287995491"],"abstract_inverted_index":{"We":[0,277],"propose":[1],"scene":[2,103,256],"graph":[3,106,257],"auto-encoder":[4,206],"(SGAE)":[5],"that":[6,149,208],"incorporates":[7],"the":[8,13,26,42,69,86,102,122,136,154,166,171,181,188,198,201,205,210,217,224,234,237,241,251,255,259,264,268,279,284,301,311,325,347,353],"language":[9,80,137,168,202,218,339],"inductive":[10,27,219,269,335],"bias":[11,28,77,270,336],"into":[12],"encoder-decoder":[14,88,211,235,242],"image":[15,128,213,287,343],"captioning":[16,288],"framework":[17],"for":[18],"more":[19,96,331],"human-like":[20,248],"captions.":[21,98],"Intuitively,":[22],"we":[23,40,92,100,139,184,323],"humans":[24,93],"use":[25,101,140],"to":[29,52,84,142,174,207,215,233,243,247,254,318],"compose":[30],"collocations":[31,232],"and":[32,57,94,118,131,170,196,236,263,304,341],"contextual":[33],"inferences":[34],"in":[35,153,180,187,275,361],"discourse.":[36],"For":[37],"example,":[38],"when":[39,68],"see":[41],"relation":[43],"\"a":[44,59],"person":[45,60],"on":[46,64,283,300,310,329],"a":[47,62,65,79,144,178,295,305],"bike\",":[48],"it":[49],"is":[50,71,82,113,271,315,359],"natural":[51],"replace":[53],"\"on\"":[54],"with":[55],"\"ride\"":[56],"infer":[58],"riding":[61],"bike":[63],"road\"":[66],"even":[67,316],"\"road\"":[70],"not":[72],"evident.":[73],"Therefore,":[74],"exploiting":[75],"such":[76,177],"as":[78,91,250],"prior":[81,169],"expected":[83],"help":[85],"conventional":[87],"models":[89],"reason":[90],"generate":[95],"descriptive":[97,231],"Specifically,":[99],"graph-a":[104],"directed":[105],"(":[107,129,133,147],"G)":[108],"where":[109,163,290],"an":[110],"object":[111],"node":[112],"connected":[114],"by":[115],"adjective":[116],"nodes":[117],"relationship":[119],"nodes-to":[120],"represent":[121],"complex":[123],"structural":[124],"layout":[125],"of":[126,200,204,209,281,327,349,355],"both":[127,350],"I)":[130],"sentence":[132],"S).":[134],"In":[135,221],"domain,":[138,183],"SGAE":[141,282,328],"learn":[143],"dictionary":[145,261],"set":[146],"D)":[148],"helps":[150],"reconstruct":[151],"sentences":[152],"S\u2192":[155],"G<sub>S</sub>":[156],"\u2192":[157,159,191,193],"D":[158,164,186,192,226],"S":[160,194],"auto-encoding":[161],"pipeline,":[162],"encodes":[165],"desired":[167],"decoder":[172,203],"learns":[173],"caption":[175],"from":[176,337],"prior;":[179],"vision-language":[182],"share":[185],"I\u2192":[189],"G<sub>I</sub>":[190],"pipeline":[195],"distill":[197],"knowledge":[199],"based":[212],"captioner":[214],"transfer":[216],"bias.":[220],"this":[222],"way,":[223],"shared":[225,260],"provides":[227],"hidden":[228],"embeddings":[229,246],"about":[230],"distillation":[238],"strategy":[239],"teaches":[240],"transform":[244],"these":[245],"captions":[249],"auto-encoder.":[252],"Thanks":[253],"representation,":[258],"set,":[262],"Knowledge":[265],"Distillation":[266],"strategy,":[267],"transferred":[272],"across":[273],"domains":[274],"principle.":[276],"validate":[278,324],"effectiveness":[280],"challenging":[285,332],"MS-COCO":[286],"benchmark,":[289],"our":[291],"SGAE-based":[292],"single-model":[293],"achieves":[294],"new":[296],"state-of-the-art":[297],"129.6":[298],"CIDEr-D":[299,308],"Karpathy":[302],"split,":[303],"competitive":[306],"126.6":[307],"(c40)":[309],"official":[312],"server,":[313],"which":[314],"comparable":[317],"other":[319,338],"ensemble":[320],"models.":[321],"Furthermore,":[322],"transferability":[326],"two":[330],"settings:":[333],"transferring":[334],"corpora":[340],"unpaired":[342],"captioning.":[344],"Once":[345],"again,":[346],"results":[348],"settings":[351],"confirm":[352],"superiority":[354],"SGAE.":[356],"The":[357],"code":[358],"released":[360],"https://github.com/yangxuntu/SGAE.":[362]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":12},{"year":2024,"cited_by_count":14},{"year":2023,"cited_by_count":15},{"year":2022,"cited_by_count":18},{"year":2021,"cited_by_count":5},{"year":2020,"cited_by_count":1}],"updated_date":"2026-04-05T17:49:38.594831","created_date":"2025-10-10T00:00:00"}