{"id":"https://openalex.org/W4417098694","doi":"https://doi.org/10.18653/v1/2026.findings-acl.814","title":"Large-Scale Diverse Synthesis for Mid-Training","display_name":"Large-Scale Diverse Synthesis for Mid-Training","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W4417098694","doi":"https://doi.org/10.18653/v1/2026.findings-acl.814"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.814","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.814","pdf_url":"https://aclanthology.org/2026.findings-acl.814.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":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.814.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5034095019","display_name":"X.P. Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xuemiao Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063924054","display_name":"Ching\u2010Ting Tu","orcid":"https://orcid.org/0000-0003-2410-726X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chengying Tu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Can Ren","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Can Ren","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081667345","display_name":"Rongxiang Weng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rongxiang Weng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100934344","display_name":"Hongfei Yan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hongfei Yan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100695186","display_name":"Jingang Wang","orcid":"https://orcid.org/0000-0003-4221-2629"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jingang Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5114353815","display_name":"Xunliang Cai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xunliang Cai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.03269424,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"16517","last_page":"16539"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.6319000124931335,"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":0.6319000124931335,"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":0.1005999967455864,"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":0.07450000196695328,"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/scalability","display_name":"Scalability","score":0.7360000014305115},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.694100022315979},{"id":"https://openalex.org/keywords/annotation","display_name":"Annotation","score":0.550000011920929},{"id":"https://openalex.org/keywords/synthetic-data","display_name":"Synthetic data","score":0.483599990606308},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.4636000096797943},{"id":"https://openalex.org/keywords/data-curation","display_name":"Data curation","score":0.4259999990463257},{"id":"https://openalex.org/keywords/data-modeling","display_name":"Data modeling","score":0.3691999912261963},{"id":"https://openalex.org/keywords/resource","display_name":"Resource (disambiguation)","score":0.3587000072002411},{"id":"https://openalex.org/keywords/face","display_name":"Face (sociological concept)","score":0.34779998660087585}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7432000041007996},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7360000014305115},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.694100022315979},{"id":"https://openalex.org/C2776321320","wikidata":"https://www.wikidata.org/wiki/Q857525","display_name":"Annotation","level":2,"score":0.550000011920929},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.483599990606308},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.4636000096797943},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4374000132083893},{"id":"https://openalex.org/C91632574","wikidata":"https://www.wikidata.org/wiki/Q15088675","display_name":"Data curation","level":2,"score":0.4259999990463257},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.40709999203681946},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.3691999912261963},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.3587000072002411},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.3544999957084656},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.34779998660087585},{"id":"https://openalex.org/C75684735","wikidata":"https://www.wikidata.org/wiki/Q858810","display_name":"Big data","level":2,"score":0.33219999074935913},{"id":"https://openalex.org/C109747225","wikidata":"https://www.wikidata.org/wiki/Q815758","display_name":"Scarcity","level":2,"score":0.32850000262260437},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.328000009059906},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3093000054359436},{"id":"https://openalex.org/C62230096","wikidata":"https://www.wikidata.org/wiki/Q275969","display_name":"Crowdsourcing","level":2,"score":0.3077999949455261},{"id":"https://openalex.org/C2778348673","wikidata":"https://www.wikidata.org/wiki/Q739302","display_name":"Production (economics)","level":2,"score":0.3061999976634979},{"id":"https://openalex.org/C72634772","wikidata":"https://www.wikidata.org/wiki/Q386824","display_name":"Data integration","level":2,"score":0.29440000653266907},{"id":"https://openalex.org/C2778915421","wikidata":"https://www.wikidata.org/wiki/Q3643177","display_name":"Performance improvement","level":2,"score":0.289900004863739},{"id":"https://openalex.org/C2777601683","wikidata":"https://www.wikidata.org/wiki/Q6499736","display_name":"Vocabulary","level":2,"score":0.28870001435279846},{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.2849000096321106},{"id":"https://openalex.org/C163985040","wikidata":"https://www.wikidata.org/wiki/Q1172399","display_name":"Data acquisition","level":2,"score":0.28439998626708984},{"id":"https://openalex.org/C2781316041","wikidata":"https://www.wikidata.org/wiki/Q1230584","display_name":"Diversity (politics)","level":2,"score":0.27720001339912415},{"id":"https://openalex.org/C2777220311","wikidata":"https://www.wikidata.org/wiki/Q6423340","display_name":"Knowledge acquisition","level":2,"score":0.2759999930858612},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.265500009059906},{"id":"https://openalex.org/C1668388","wikidata":"https://www.wikidata.org/wiki/Q1149776","display_name":"Data management","level":2,"score":0.2639000117778778},{"id":"https://openalex.org/C56288433","wikidata":"https://www.wikidata.org/wiki/Q58673","display_name":"Data manipulation language","level":2,"score":0.26190000772476196},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.2517000138759613},{"id":"https://openalex.org/C2983685735","wikidata":"https://www.wikidata.org/wiki/Q5227355","display_name":"Data source","level":2,"score":0.25029999017715454}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.814","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.814","pdf_url":"https://aclanthology.org/2026.findings-acl.814.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":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2508.01326","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.01326","pdf_url":"https://arxiv.org/pdf/2508.01326","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"doi:10.48550/arxiv.2508.01326","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2508.01326","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.814","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.814","pdf_url":"https://aclanthology.org/2026.findings-acl.814.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":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4417098694.pdf","grobid_xml":"https://content.openalex.org/works/W4417098694.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Mid-training":[0],"has":[1],"become":[2],"critical":[3],"for":[4,57,64],"enhancing":[5],"the":[6,18,65,118],"knowledge":[7],"and":[8,31,38,53,68,115,156],"reasoning":[9],"ability":[10],"of":[11,20,111,140],"large":[12],"language":[13],"models":[14,139],"(LLMs),":[15],"especially":[16],"through":[17],"utilization":[19],"large-scale":[21],"synthetic":[22],"data.However,":[23],"existing":[24],"data":[25,56,75,105,154],"synthesis":[26,72,81],"methods":[27],"often":[28],"generate":[29],"simplistic":[30],"homogeneous":[32],"QA":[33,55],"pairs,":[34],"with":[35,147],"limited":[36],"scale":[37],"diversity.To":[39],"address":[40],"this,":[41],"we":[42],"propose":[43],"BoostQA,":[44],"a":[45,107],"novel":[46],"framework":[47],"designed":[48],"to":[49,73,82,90],"synthesize":[50],"large-scale,":[51],"diverse,":[52],"high-quality":[54],"midtraining.BoostQA":[58],"introduces":[59],"model":[60,130,152],"probes":[61],"during":[62],"mid-training":[63,96,121],"first":[66],"time":[67],"implements":[69],"STEM-focused":[70],"multi-grade":[71],"boost":[74],"diversity":[76],"as":[77,79,151],"well":[78],"high-difficulty":[80],"alleviate":[83],"difficulty":[84],"degradation,":[85],"followed":[86],"by":[87,95],"answer":[88],"refinement":[89],"further":[91],"improve":[92],"quality.Extensive":[93],"experiments":[94],"Llama-3":[97],"8B":[98],"demonstrate":[99],"that":[100],"using":[101],"only":[102],"20Btoken":[103],"BoostQA":[104,127],"achieves":[106,131],"significant":[108],"average":[109,133],"improvement":[110],"12.74%":[112],"on":[113,122],"MMLU":[114],"CMMLU":[116],"over":[117],"pre-training":[119],"baseline.After":[120],"500B":[123],"tokens,":[124],"including":[125],"100Btoken":[126],"data,":[128],"our":[129],"SOTA":[132],"results":[134],"across":[135],"benchmarks":[136],"among":[137],"mainstream":[138],"comparable":[141],"size.BoostQA":[142],"also":[143],"demonstrates":[144],"robust":[145],"scalability,":[146],"performance":[148],"consistently":[149],"improving":[150],"size,":[153],"volume,":[155],"initial":[157],"FLOPs":[158],"scale.":[159]},"counts_by_year":[],"updated_date":"2026-08-05T07:39:15.569665","created_date":"2025-10-10T00:00:00"}