{"id":"https://openalex.org/W4296069291","doi":"https://doi.org/10.21437/interspeech.2022-10858","title":"Extending Compositional Attention Networks for Social Reasoning in Videos","display_name":"Extending Compositional Attention Networks for Social Reasoning in Videos","publication_year":2022,"publication_date":"2022-09-16","ids":{"openalex":"https://openalex.org/W4296069291","doi":"https://doi.org/10.21437/interspeech.2022-10858"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2022-10858","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2022-10858","pdf_url":null,"source":{"id":"https://openalex.org/S4363604309","display_name":"Interspeech 2022","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2022","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2210.01191","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5001431318","display_name":"Christina Sartzetaki","orcid":"https://orcid.org/0009-0009-2057-5614"},"institutions":[{"id":"https://openalex.org/I174458059","display_name":"National Technical University of Athens","ror":"https://ror.org/03cx6bg69","country_code":"GR","type":"education","lineage":["https://openalex.org/I174458059"]}],"countries":["GR"],"is_corresponding":true,"raw_author_name":"Christina Sartzetaki","raw_affiliation_strings":["School of ECE, National Technical University of Athens, Greece"],"affiliations":[{"raw_affiliation_string":"School of ECE, National Technical University of Athens, Greece","institution_ids":["https://openalex.org/I174458059"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026032406","display_name":"Georgios Paraskevopoulos","orcid":"https://orcid.org/0000-0003-4067-6294"},"institutions":[{"id":"https://openalex.org/I4210098364","display_name":"Institute for Language and Speech Processing","ror":"https://ror.org/00z24kr14","country_code":"GR","type":"nonprofit","lineage":["https://openalex.org/I4210098364"]},{"id":"https://openalex.org/I174458059","display_name":"National Technical University of Athens","ror":"https://ror.org/03cx6bg69","country_code":"GR","type":"education","lineage":["https://openalex.org/I174458059"]}],"countries":["GR"],"is_corresponding":false,"raw_author_name":"Georgios Paraskevopoulos","raw_affiliation_strings":["Institute for Speech and Language Processing, Athens, Greece","School of ECE, National Technical University of Athens, Greece"],"affiliations":[{"raw_affiliation_string":"Institute for Speech and Language Processing, Athens, Greece","institution_ids":["https://openalex.org/I4210098364"]},{"raw_affiliation_string":"School of ECE, National Technical University of Athens, Greece","institution_ids":["https://openalex.org/I174458059"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5084949286","display_name":"Alexandros Potamianos","orcid":"https://orcid.org/0009-0007-1532-5288"},"institutions":[{"id":"https://openalex.org/I174458059","display_name":"National Technical University of Athens","ror":"https://ror.org/03cx6bg69","country_code":"GR","type":"education","lineage":["https://openalex.org/I174458059"]}],"countries":["GR"],"is_corresponding":false,"raw_author_name":"Alexandros Potamianos","raw_affiliation_strings":["School of ECE, National Technical University of Athens, Greece"],"affiliations":[{"raw_affiliation_string":"School of ECE, National Technical University of Athens, Greece","institution_ids":["https://openalex.org/I174458059"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5001431318"],"corresponding_institution_ids":["https://openalex.org/I174458059"],"apc_list":null,"apc_paid":null,"fwci":0.1196,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.41064585,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"1116","last_page":"1120"},"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9984999895095825,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9934999942779541,"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":0.6340376138687134},{"id":"https://openalex.org/keywords/cognitive-science","display_name":"Cognitive science","score":0.40374845266342163},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3240722417831421},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.1723904013633728}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6340376138687134},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.40374845266342163},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3240722417831421},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.1723904013633728}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.21437/interspeech.2022-10858","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2022-10858","pdf_url":null,"source":{"id":"https://openalex.org/S4363604309","display_name":"Interspeech 2022","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2022","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2210.01191","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2210.01191","pdf_url":"https://arxiv.org/pdf/2210.01191","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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2210.01191","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2210.01191","pdf_url":"https://arxiv.org/pdf/2210.01191","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"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":30,"referenced_works":["https://openalex.org/W1933349210","https://openalex.org/W2131494463","https://openalex.org/W2293453011","https://openalex.org/W2561715562","https://openalex.org/W2606982687","https://openalex.org/W2619383789","https://openalex.org/W2745461083","https://openalex.org/W2787581402","https://openalex.org/W2799244840","https://openalex.org/W2904452845","https://openalex.org/W2953388933","https://openalex.org/W2962949233","https://openalex.org/W2963518342","https://openalex.org/W2963541336","https://openalex.org/W2963644680","https://openalex.org/W2964010806","https://openalex.org/W2964118342","https://openalex.org/W2964306921","https://openalex.org/W2966683369","https://openalex.org/W3015965768","https://openalex.org/W3016211260","https://openalex.org/W3020743079","https://openalex.org/W3089358914","https://openalex.org/W3093534251","https://openalex.org/W3104219743","https://openalex.org/W3111321733","https://openalex.org/W3187433838","https://openalex.org/W3212002421","https://openalex.org/W4288286281","https://openalex.org/W4297749157"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052","https://openalex.org/W4402327032","https://openalex.org/W2382290278"],"abstract_inverted_index":{"We":[0,16,62,95],"propose":[1,28],"a":[2,29,37,58,113],"novel":[3],"deep":[4],"architecture":[5,84],"for":[6,68],"the":[7,18,81,99,107,123],"task":[8,100],"of":[9,22,45,57,101,119],"reasoning":[10,20,53],"about":[11],"social":[12],"interactions":[13],"in":[14,72,106,117],"videos.":[15],"leverage":[17,87],"multi-step":[19],"capabilities":[21],"Compositional":[23],"Attention":[24],"Networks":[25],"(MAC),":[26],"and":[27,111],"multimodal":[30,88],"extension":[31],"(MAC-X).":[32],"MAC-X":[33,65,83,97],"is":[34],"based":[35],"on":[36],"recurrent":[38],"cell":[39],"that":[40,80],"performs":[41],"iterative":[42],"mid-level":[43,92],"fusion":[44,93],"input":[46,70,89],"modalities":[47],"(visual,":[48],"auditory,":[49],"text)":[50],"over":[51,122],"multiple":[52],"steps,":[54],"by":[55],"use":[56],"temporal":[59,69],"attention":[60],"mechanism.":[61],"then":[63],"combine":[64],"with":[66],"LSTMs":[67],"processing":[71],"an":[73],"end-to-end":[74],"architecture.":[75],"Our":[76],"ablation":[77],"studies":[78],"show":[79],"proposed":[82],"can":[85],"effectively":[86],"cues":[90],"using":[91],"mechanisms.":[94],"apply":[96],"to":[98],"Social":[102,108],"Video":[103],"Question":[104],"Answering":[105],"IQ":[109],"dataset":[110],"obtain":[112],"2.5%":[114],"absolute":[115],"improvement":[116],"terms":[118],"binary":[120],"accuracy":[121],"current":[124],"state-of-the-art.":[125]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2023,"cited_by_count":1}],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2022-09-17T00:00:00"}