{"id":"https://openalex.org/W2948902814","doi":"https://doi.org/10.18653/v1/w19-4801","title":"Transcoding Compositionally: Using Attention to Find More Generalizable Solutions","display_name":"Transcoding Compositionally: Using Attention to Find More Generalizable Solutions","publication_year":2019,"publication_date":"2019-01-01","ids":{"openalex":"https://openalex.org/W2948902814","doi":"https://doi.org/10.18653/v1/w19-4801","mag":"2948902814"},"language":"en","primary_location":{"id":"doi:10.18653/v1/w19-4801","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/w19-4801","pdf_url":"https://www.aclweb.org/anthology/W19-4801.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2019 ACL Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.aclweb.org/anthology/W19-4801.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5042961025","display_name":"Kris Korrel","orcid":null},"institutions":[{"id":"https://openalex.org/I887064364","display_name":"University of Amsterdam","ror":"https://ror.org/04dkp9463","country_code":"NL","type":"education","lineage":["https://openalex.org/I887064364"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Kris Korrel","raw_affiliation_strings":["University of Amsterdam"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Amsterdam","institution_ids":["https://openalex.org/I887064364"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018069499","display_name":"Dieuwke Hupkes","orcid":null},"institutions":[{"id":"https://openalex.org/I887064364","display_name":"University of Amsterdam","ror":"https://ror.org/04dkp9463","country_code":"NL","type":"education","lineage":["https://openalex.org/I887064364"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Dieuwke Hupkes","raw_affiliation_strings":["University of Amsterdam"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Amsterdam","institution_ids":["https://openalex.org/I887064364"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074102828","display_name":"Verna Dankers","orcid":null},"institutions":[{"id":"https://openalex.org/I887064364","display_name":"University of Amsterdam","ror":"https://ror.org/04dkp9463","country_code":"NL","type":"education","lineage":["https://openalex.org/I887064364"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Verna Dankers","raw_affiliation_strings":["University of Amsterdam"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Amsterdam","institution_ids":["https://openalex.org/I887064364"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5007465102","display_name":"Elia Bruni","orcid":null},"institutions":[{"id":"https://openalex.org/I170486558","display_name":"Universitat Pompeu Fabra","ror":"https://ror.org/04n0g0b29","country_code":"ES","type":"education","lineage":["https://openalex.org/I170486558"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Elia Bruni","raw_affiliation_strings":["Universitat Pompeu Fabra"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Universitat Pompeu Fabra","institution_ids":["https://openalex.org/I170486558"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"11"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","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"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","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/T10181","display_name":"Natural Language Processing Techniques","score":0.9998000264167786,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9991999864578247,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.8124154806137085},{"id":"https://openalex.org/keywords/principle-of-compositionality","display_name":"Principle of compositionality","score":0.805570662021637},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7941238880157471},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.7209066152572632},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.715335488319397},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.5905267000198364},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5048416256904602},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.4503421485424042},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.44866806268692017},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.35745012760162354},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.18463638424873352},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.10218608379364014}],"concepts":[{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.8124154806137085},{"id":"https://openalex.org/C121375916","wikidata":"https://www.wikidata.org/wiki/Q936559","display_name":"Principle of compositionality","level":2,"score":0.805570662021637},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7941238880157471},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.7209066152572632},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.715335488319397},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.5905267000198364},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5048416256904602},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.4503421485424042},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44866806268692017},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.35745012760162354},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.18463638424873352},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.10218608379364014},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C54355233","wikidata":"https://www.wikidata.org/wiki/Q7162","display_name":"Genetics","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0}],"mesh":[],"locations_count":7,"locations":[{"id":"doi:10.18653/v1/w19-4801","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/w19-4801","pdf_url":"https://www.aclweb.org/anthology/W19-4801.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2019 ACL Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:1906.01234","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1906.01234","pdf_url":"https://arxiv.org/pdf/1906.01234","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:oai:dare.uva.nl:publications/01210a30-cf84-41ea-8bbc-66dcd470a739","is_oa":true,"landing_page_url":"https://handle.uba.uva.nl/personal/pure/en/publications/transcoding-compositionally-using-attention-to-find-more-generalizable-solutions(01210a30-cf84-41ea-8bbc-66dcd470a739).html","pdf_url":"https://pure.uva.nl/ws/files/49860836/W19_4801.pdf","source":{"id":"https://openalex.org/S4306400088","display_name":"UvA-DARE (University of Amsterdam)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I887064364","host_organization_name":"University of Amsterdam","host_organization_lineage":["https://openalex.org/I887064364"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Korrel, K, Hupkes, D, Dankers, V & Bruni, E 2019, Transcoding compositionally: using attention to find more generalizable solutions. in T Linzen, G Chrupa\u0142a, Y Belinkov & D Hupkes (eds), The BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP at ACL 2019 : ACL 2019 : proceedings of the Second Workshop : August 1, 2019, Florence, Italy. Stroudsburg, PA, pp. 1-11, BlackboxNLP 2019, 1/08/19. https://doi.org/10.18653/v1/W19-4801","raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"pmh:uvapub:oai:dare.uva.nl:publications/01210a30-cf84-41ea-8bbc-66dcd470a739","is_oa":true,"landing_page_url":"https://dare.uva.nl/personal/pure/en/publications/transcoding-compositionally-using-attention-to-find-more-generalizable-solutions(01210a30-cf84-41ea-8bbc-66dcd470a739).html","pdf_url":"https://dare.uva.nl/personal/pure/en/publications/transcoding-compositionally-using-attention-to-find-more-generalizable-solutions(01210a30-cf84-41ea-8bbc-66dcd470a739).html","source":{"id":"https://openalex.org/S4306401843","display_name":"Data Archiving and Networked Services (DANS)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1322597698","host_organization_name":"Royal Netherlands Academy of Arts and Sciences","host_organization_lineage":["https://openalex.org/I1322597698"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"The BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP at ACL 2019: ACL 2019 : proceedings of the Second Workshop : August 1, 2019, Florence, Italy, 1 - 11","raw_type":"info:eu-repo/semantics/conferencepaper"},{"id":"pmh:oai:dare.uva.nl:openaire/01210a30-cf84-41ea-8bbc-66dcd470a739","is_oa":true,"landing_page_url":"https://hdl.handle.net/11245.1/01210a30-cf84-41ea-8bbc-66dcd470a739","pdf_url":null,"source":{"id":"https://openalex.org/S4306400088","display_name":"UvA-DARE (University of Amsterdam)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I887064364","host_organization_name":"University of Amsterdam","host_organization_lineage":["https://openalex.org/I887064364"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Korrel, K, Hupkes, D, Dankers, V & Bruni, E 2019, Transcoding compositionally: using attention to find more generalizable solutions. in T Linzen, G Chrupa\u0142a, Y Belinkov & D Hupkes (eds), The BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP at ACL 2019 : ACL 2019 : proceedings of the Second Workshop : August 1, 2019, Florence, Italy. Stroudsburg, PA, pp. 1-11, BlackboxNLP 2019, 1/08/19. https://doi.org/10.18653/v1/W19-4801","raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"doi:10.48550/arxiv.1906.01234","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1906.01234","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"},{"id":"mag:2948902814","is_oa":false,"landing_page_url":null,"pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":null}],"best_oa_location":{"id":"doi:10.18653/v1/w19-4801","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/w19-4801","pdf_url":"https://www.aclweb.org/anthology/W19-4801.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2019 ACL Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.6899999976158142,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[{"id":"https://openalex.org/G2084304408","display_name":null,"funder_award_id":"024.001.006","funder_id":"https://openalex.org/F4320321800","funder_display_name":"Nederlandse Organisatie voor Wetenschappelijk Onderzoek"},{"id":"https://openalex.org/G4393542428","display_name":"Multimodal Agents Grounded via Interactive Communication","funder_award_id":"790369","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G6728881174","display_name":null,"funder_award_id":"Gravitation Grant 024.001.006","funder_id":"https://openalex.org/F4320321800","funder_display_name":"Nederlandse Organisatie voor Wetenschappelijk Onderzoek"}],"funders":[{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"},{"id":"https://openalex.org/F4320321800","display_name":"Nederlandse Organisatie voor Wetenschappelijk Onderzoek","ror":"https://ror.org/04jsz6e67"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2948902814.pdf","grobid_xml":"https://content.openalex.org/works/W2948902814.grobid-xml"},"referenced_works_count":25,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W2116723809","https://openalex.org/W2121029939","https://openalex.org/W2130942839","https://openalex.org/W2157331557","https://openalex.org/W2194321275","https://openalex.org/W2325237720","https://openalex.org/W2547875792","https://openalex.org/W2548228487","https://openalex.org/W2561715562","https://openalex.org/W2613904329","https://openalex.org/W2773570800","https://openalex.org/W2786209943","https://openalex.org/W2788751659","https://openalex.org/W2804339109","https://openalex.org/W2889947987","https://openalex.org/W2950202163","https://openalex.org/W2950820547","https://openalex.org/W2952718809","https://openalex.org/W2962816513","https://openalex.org/W2963123301","https://openalex.org/W2963267799","https://openalex.org/W2963403868","https://openalex.org/W2963430224","https://openalex.org/W2972778456"],"related_works":["https://openalex.org/W3093630174","https://openalex.org/W3202010866","https://openalex.org/W2939413764","https://openalex.org/W2963914915","https://openalex.org/W3037664376","https://openalex.org/W3092967498","https://openalex.org/W2979065840","https://openalex.org/W2952280909","https://openalex.org/W3213371964","https://openalex.org/W2900556903","https://openalex.org/W2250473257","https://openalex.org/W3132426319","https://openalex.org/W3008568373","https://openalex.org/W2766742239","https://openalex.org/W3006583220","https://openalex.org/W2943085977","https://openalex.org/W2952349142","https://openalex.org/W1644623430","https://openalex.org/W2949794029","https://openalex.org/W3207039423"],"abstract_inverted_index":{"While":[0],"sequence-to-sequence":[1,163],"models":[2],"have":[3],"shown":[4],"remarkable":[5],"generalization":[6],"power":[7],"across":[8],"several":[9],"natural":[10],"language":[11],"tasks,":[12],"their":[13],"construct":[14],"of":[15,55,95,110,114],"solutions":[16,99,115],"are":[17,59,87,104,117],"argued":[18],"to":[19,39,42,90,128,133,140,156],"be":[20],"less":[21],"compositional":[22,44,93],"than":[23,160],"human-like":[24],"generalization.":[25],"In":[26,49],"this":[27,126],"paper,":[28],"we":[29],"present":[30],"seq2attn,":[31,50],"a":[32,62,130,142,157,161],"new":[33,131],"architecture":[34],"that":[35,64,73,116,136,151],"is":[36],"specifically":[37,88],"designed":[38],"exploit":[40,125],"attention":[41],"find":[43],"patterns":[45],"in":[46],"the":[47,51,66,92,102,112,138],"input.":[48],"two":[52,84],"standard":[53,162],"components":[54],"an":[56],"encoder-decoder":[57],"model":[58,103,143],"connected":[60],"via":[61],"transcoder,":[63],"modulates":[65],"information":[67],"flow":[68],"between":[69],"them.":[70],"We":[71,124,149],"show":[72,150],"seq2attn":[74,152],"can":[75],"successfully":[76],"generalize,":[77],"without":[78],"requiring":[79],"any":[80],"additional":[81],"supervision,":[82],"on":[83],"tasks":[85],"which":[86,141],"constructed":[89],"challenge":[91],"skills":[94],"neural":[96],"networks.":[97],"The":[98],"found":[100,118],"by":[101],"highly":[105],"interpretable,":[106],"allowing":[107],"easy":[108],"analysis":[109],"both":[111],"types":[113],"and":[119],"potential":[120],"causes":[121],"for":[122],"mistakes.":[123],"opportunity":[127],"introduce":[129],"paradigm":[132],"test":[134],"compositionality":[135],"studies":[137],"extent":[139],"overgeneralizes":[144],"when":[145],"confronted":[146],"with":[147],"exceptions.":[148],"exhibits":[153],"such":[154],"overgeneralization":[155],"larger":[158],"degree":[159],"model.":[164]},"counts_by_year":[{"year":2020,"cited_by_count":1},{"year":2019,"cited_by_count":1}],"updated_date":"2026-07-20T07:56:41.581041","created_date":"2022-07-29T00:00:00"}