{"id":"https://openalex.org/W4224312921","doi":"https://doi.org/10.1145/3485447.3512021","title":"Can Small Heads Help? Understanding and Improving Multi-Task Generalization","display_name":"Can Small Heads Help? Understanding and Improving Multi-Task Generalization","publication_year":2022,"publication_date":"2022-04-25","ids":{"openalex":"https://openalex.org/W4224312921","doi":"https://doi.org/10.1145/3485447.3512021"},"language":"en","primary_location":{"id":"doi:10.1145/3485447.3512021","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3485447.3512021","pdf_url":null,"source":{"id":"https://openalex.org/S4363608783","display_name":"Proceedings of the ACM Web Conference 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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM Web Conference 2022","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.1145/3485447.3512021","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100677212","display_name":"Yuyan Wang","orcid":"https://orcid.org/0000-0001-5585-3789"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Yuyan Wang","raw_affiliation_strings":["Google Research, USA"],"affiliations":[{"raw_affiliation_string":"Google Research, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100631150","display_name":"Zhe Zhao","orcid":"https://orcid.org/0000-0002-6847-0186"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhe Zhao","raw_affiliation_strings":["Google Research, USA"],"affiliations":[{"raw_affiliation_string":"Google Research, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101990493","display_name":"Bo Dai","orcid":"https://orcid.org/0000-0003-0777-9232"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bo Dai","raw_affiliation_strings":["Google Research, USA"],"affiliations":[{"raw_affiliation_string":"Google Research, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021315343","display_name":"Christopher Fifty","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Christopher Fifty","raw_affiliation_strings":["Google Research, USA"],"affiliations":[{"raw_affiliation_string":"Google Research, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100702458","display_name":"Dong Lin","orcid":"https://orcid.org/0000-0001-6847-4024"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dong Lin","raw_affiliation_strings":["Google Research, USA"],"affiliations":[{"raw_affiliation_string":"Google Research, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079085366","display_name":"Lichan Hong","orcid":"https://orcid.org/0009-0004-9563-554X"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Lichan Hong","raw_affiliation_strings":["Google Research, USA"],"affiliations":[{"raw_affiliation_string":"Google Research, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100318291","display_name":"Wei Li","orcid":"https://orcid.org/0000-0002-8278-1765"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Li Wei","raw_affiliation_strings":["Google Research, USA"],"affiliations":[{"raw_affiliation_string":"Google Research, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5028125399","display_name":"Ed H.","orcid":"https://orcid.org/0000-0003-3230-5338"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ed H. Chi","raw_affiliation_strings":["Google Research, USA"],"affiliations":[{"raw_affiliation_string":"Google Research, USA","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5100677212"],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":0.8315,"has_fulltext":false,"cited_by_count":8,"citation_normalized_percentile":{"value":0.72979961,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"3009","last_page":"3019"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.9898999929428101,"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.9898999929428101,"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":0.9882000088691711,"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/T12031","display_name":"Speech and dialogue systems","score":0.9779000282287598,"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":0.7632927298545837},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.7122876644134521},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6569334268569946},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.362338125705719},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.08819335699081421},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.06951487064361572},{"id":"https://openalex.org/keywords/systems-engineering","display_name":"Systems engineering","score":0.05171605944633484}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7632927298545837},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.7122876644134521},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6569334268569946},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.362338125705719},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08819335699081421},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.06951487064361572},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.05171605944633484},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3485447.3512021","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3485447.3512021","pdf_url":null,"source":{"id":"https://openalex.org/S4363608783","display_name":"Proceedings of the ACM Web Conference 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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM Web Conference 2022","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3485447.3512021","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3485447.3512021","pdf_url":null,"source":{"id":"https://openalex.org/S4363608783","display_name":"Proceedings of the ACM Web Conference 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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM Web Conference 2022","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":28,"referenced_works":["https://openalex.org/W207651325","https://openalex.org/W1536680647","https://openalex.org/W1607093780","https://openalex.org/W1896424170","https://openalex.org/W1970031898","https://openalex.org/W1977310087","https://openalex.org/W2060727197","https://openalex.org/W2112796928","https://openalex.org/W2117130368","https://openalex.org/W2162888803","https://openalex.org/W2219888463","https://openalex.org/W2510317721","https://openalex.org/W2512971201","https://openalex.org/W2549401308","https://openalex.org/W2556468274","https://openalex.org/W2740941491","https://openalex.org/W2802620382","https://openalex.org/W2809290718","https://openalex.org/W2903852246","https://openalex.org/W2912127016","https://openalex.org/W2963706742","https://openalex.org/W2963877604","https://openalex.org/W2973172293","https://openalex.org/W3099732023","https://openalex.org/W3117945726","https://openalex.org/W4234552385","https://openalex.org/W6620434334","https://openalex.org/W6734815144"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W3162204513","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890"],"abstract_inverted_index":{"Multi-task":[0],"learning":[1,7,25,63,116],"aims":[2],"to":[3,33],"solve":[4],"multiple":[5],"machine":[6],"tasks":[8],"at":[9],"the":[10,41,53,69,73,85,119,124,133,136],"same":[11],"time,":[12],"with":[13],"good":[14],"solutions":[15],"being":[16],"both":[17],"generalizable":[18],"and":[19,37,64,90,131],"Pareto":[20],"optimal.":[21],"A":[22],"multi-task":[23,62,66],"deep":[24],"model":[26,125],"consists":[27],"of":[28,43,72,129,135],"a":[29,95,108,114],"shared":[30,74,91,137],"representation":[31,92,115],"learned":[32],"capture":[34],"task":[35,58,105],"commonalities,":[36],"task-specific":[38,120],"sub-networks":[39,121],"capturing":[40],"specificities":[42],"each":[44],"task.":[45],"In":[46],"this":[47],"work,":[48],"we":[49],"offer":[50],"insights":[51],"on":[52],"under-explored":[54],"trade-off":[55,80],"between":[56,87],"minimizing":[57],"training":[59],"conflicts":[60,106],"in":[61,107],"improving":[65],"generalization,":[67],"i.e.":[68],"generalization":[70],"capability":[71],"presentation":[75],"across":[76],"all":[77],"tasks.":[78],"The":[79],"can":[81],"be":[82],"viewed":[83],"as":[84],"tension":[86],"multi-objective":[88,96],"optimization":[89,97],"learning:":[93],"As":[94],"problem,":[98],"sufficient":[99],"parameterization":[100],"is":[101],"needed":[102],"for":[103],"mitigating":[104],"constrained":[109],"solution":[110],"space;":[111],"However,":[112],"from":[113],"perspective,":[117],"over-parameterizing":[118],"may":[122],"give":[123],"too":[126],"many":[127],"\u201ddegrees":[128],"freedom\u201d":[130],"impedes":[132],"generalizability":[134],"representation.":[138]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":4}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}