{"id":"https://openalex.org/W4412877069","doi":"https://doi.org/10.1145/3711896.3737117","title":"Scaling Transformers for Discriminative Recommendation via Generative Pretraining","display_name":"Scaling Transformers for Discriminative Recommendation via Generative Pretraining","publication_year":2025,"publication_date":"2025-08-03","ids":{"openalex":"https://openalex.org/W4412877069","doi":"https://doi.org/10.1145/3711896.3737117"},"language":"en","primary_location":{"id":"doi:10.1145/3711896.3737117","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3711896.3737117","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3711896.3737117","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3711896.3737117","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5024620814","display_name":"C.-L. Wang","orcid":"https://orcid.org/0000-0002-8047-0867"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chunqi Wang","raw_affiliation_strings":["Alibaba International Digital Commercial Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0002-8047-0867","affiliations":[{"raw_affiliation_string":"Alibaba International Digital Commercial Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072179387","display_name":"Bingchao Wu","orcid":"https://orcid.org/0000-0002-8164-4240"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bingchao Wu","raw_affiliation_strings":["Alibaba International Digital Commercial Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0002-8164-4240","affiliations":[{"raw_affiliation_string":"Alibaba International Digital Commercial Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zheng Chen","orcid":"https://orcid.org/0009-0001-4871-848X"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zheng Chen","raw_affiliation_strings":["Alibaba International Digital Commercial Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0009-0001-4871-848X","affiliations":[{"raw_affiliation_string":"Alibaba International Digital Commercial Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049534107","display_name":"Lei Shen","orcid":"https://orcid.org/0000-0001-5249-7037"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lei Shen","raw_affiliation_strings":["Alibaba International Digital Commercial Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0001-5249-7037","affiliations":[{"raw_affiliation_string":"Alibaba International Digital Commercial Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100382467","display_name":"Bing Wang","orcid":"https://orcid.org/0000-0001-7976-8642"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bing Wang","raw_affiliation_strings":["Alibaba International Digital Commercial Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0001-7976-8642","affiliations":[{"raw_affiliation_string":"Alibaba International Digital Commercial Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5082008486","display_name":"Xiaoyi Zeng","orcid":"https://orcid.org/0000-0002-3742-4910"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaoyi Zeng","raw_affiliation_strings":["Alibaba International Digital Commercial Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0002-3742-4910","affiliations":[{"raw_affiliation_string":"Alibaba International Digital Commercial Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I45928872"],"apc_list":null,"apc_paid":null,"fwci":2.1023,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.88897421,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"2893","last_page":"2903"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10203","display_name":"Recommender Systems and Techniques","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10203","display_name":"Recommender Systems and Techniques","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10824","display_name":"Image Retrieval and Classification Techniques","score":0.9977999925613403,"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/T11309","display_name":"Music and Audio Processing","score":0.9973999857902527,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/discriminative-model","display_name":"Discriminative model","score":0.842300295829773},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.6847000122070312},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.6002610921859741},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5938786268234253},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.4907735288143158},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.41859549283981323},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.33016806840896606},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.1394270658493042},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.09632372856140137},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.08047306537628174},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.04399314522743225}],"concepts":[{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.842300295829773},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.6847000122070312},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.6002610921859741},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5938786268234253},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.4907735288143158},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41859549283981323},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.33016806840896606},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.1394270658493042},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.09632372856140137},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.08047306537628174},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.04399314522743225},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3711896.3737117","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3711896.3737117","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3711896.3737117","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2506.03699","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.03699","pdf_url":"https://arxiv.org/pdf/2506.03699","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":"doi:10.1145/3711896.3737117","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3711896.3737117","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3711896.3737117","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","score":0.7699999809265137,"display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412877069.pdf","grobid_xml":"https://content.openalex.org/works/W4412877069.grobid-xml"},"referenced_works_count":25,"referenced_works":["https://openalex.org/W2723293840","https://openalex.org/W2783272285","https://openalex.org/W2945772520","https://openalex.org/W2962745591","https://openalex.org/W2963341956","https://openalex.org/W2963367478","https://openalex.org/W2966123616","https://openalex.org/W2984100107","https://openalex.org/W2990138404","https://openalex.org/W2994850640","https://openalex.org/W2997024057","https://openalex.org/W3034896171","https://openalex.org/W3093601757","https://openalex.org/W3106252282","https://openalex.org/W3172942063","https://openalex.org/W3197720002","https://openalex.org/W3210268599","https://openalex.org/W4306317487","https://openalex.org/W4308760184","https://openalex.org/W4382239917","https://openalex.org/W4388979610","https://openalex.org/W4389675217","https://openalex.org/W4401201922","https://openalex.org/W4403218830","https://openalex.org/W6846659131"],"related_works":["https://openalex.org/W2965546495","https://openalex.org/W4389116644","https://openalex.org/W2153315159","https://openalex.org/W3103844505","https://openalex.org/W259157601","https://openalex.org/W4396941953","https://openalex.org/W2987280934","https://openalex.org/W2404514746","https://openalex.org/W1652783584","https://openalex.org/W2082783427"],"abstract_inverted_index":{"Discriminative":[0,72],"recommendation":[1,202,219],"tasks,":[2],"such":[3],"as":[4,166,177],"CTR":[5],"(click-through":[6],"rate)":[7,11],"and":[8,58,101,115,154,204],"CVR":[9],"(conversion":[10],"prediction,":[12],"play":[13],"critical":[14],"roles":[15],"in":[16,130,146,150,214],"the":[17,55,88,120,143,158,178,195,199,208],"ranking":[18],"stage":[19],"of":[20,84,123,160,201,211],"large-scale":[21],"industrial":[22],"recommender":[23],"systems.":[24],"However,":[25],"training":[26],"a":[27,31,64,92,98,104],"discriminative":[28,99],"model":[29,60,95,147,179],"encounters":[30],"significant":[32],"overfitting":[33,41,56],"issue":[34,42,57],"induced":[35],"by":[36],"data":[37],"sparsity.":[38],"Moreover,":[39,125],"this":[40],"worsens":[43],"with":[44],"larger":[45],"models,":[46,206],"causing":[47],"them":[48],"to":[49,96,185,189,218],"underperform":[50],"smaller":[51],"ones.":[52],"To":[53],"address":[54],"enhance":[59],"scalability,":[61],"we":[62,172],"propose":[63],"framework":[65],"named":[66],"GPSD":[67,86,134],"(Generative":[68],"Pretraining":[69],"for":[70,197],"Scalable":[71],"Recommendation),":[73],"drawing":[74],"inspiration":[75],"from":[76,91,183],"generative":[77,94],"training,":[78,148],"which":[79],"exhibits":[80],"no":[81],"evident":[82],"signs":[83],"overfitting.":[85],"leverages":[87,157],"parameters":[89,181],"learned":[90],"pretrained":[93],"initialize":[97],"model,":[100],"subsequently":[102],"applies":[103],"sparse":[105],"parameter":[106],"freezing":[107],"strategy.":[108],"Extensive":[109],"experiments":[110],"conducted":[111],"on":[112],"both":[113],"industrial-scale":[114],"publicly":[116],"available":[117],"datasets":[118],"demonstrate":[119],"superior":[121],"performance":[122,164,175],"GPSD.":[124],"it":[126,140,156],"delivers":[127],"remarkable":[128],"improvements":[129,176],"online":[131],"A/B":[132],"tests.":[133],"offers":[135],"two":[136],"primary":[137],"advantages:":[138],"1)":[139],"substantially":[141],"narrows":[142],"generalization":[144],"gap":[145],"resulting":[149],"better":[151],"test":[152],"performance;":[153],"2)":[155],"scalability":[159],"Transformers,":[161],"delivering":[162],"consistent":[163,174],"gains":[165],"models":[167,203,217],"are":[168],"scaled":[169],"up.":[170],"Specifically,":[171],"observe":[173],"dense":[180],"scale":[182],"13K":[184],"0.3B,":[186],"closely":[187],"adhering":[188],"power":[190],"laws.":[191],"These":[192],"findings":[193],"pave":[194],"way":[196],"unifying":[198],"architectures":[200],"language":[205,216],"enabling":[207],"direct":[209],"application":[210],"techniques":[212],"well-established":[213],"large":[215],"models.":[220]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2025-10-10T00:00:00"}