{"id":"https://openalex.org/W4416035185","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.146","title":"Rethinking Data Selection at Scale: Random Selection is Almost All You Need","display_name":"Rethinking Data Selection at Scale: Random Selection is Almost All You Need","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416035185","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.146"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.146","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.146","pdf_url":"https://aclanthology.org/2025.findings-emnlp.146.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: EMNLP 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.146.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5083820813","display_name":"Tingyu Xia","orcid":"https://orcid.org/0000-0002-4785-9969"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tingyu Xia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100604886","display_name":"Bowen Yu","orcid":"https://orcid.org/0000-0002-6804-1859"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bowen Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014400759","display_name":"Kai Dang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kai Dang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100459622","display_name":"Yang An","orcid":"https://orcid.org/0000-0001-8838-3197"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"An Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102024536","display_name":"Yuan Wu","orcid":"https://orcid.org/0000-0001-6289-5872"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuan Wu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071435827","display_name":"Yuan Tian","orcid":"https://orcid.org/0000-0001-6073-8582"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuan Tian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101735569","display_name":"Yi Chang","orcid":"https://orcid.org/0000-0002-2417-1328"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yi Chang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100612233","display_name":"Junyang Lin","orcid":"https://orcid.org/0000-0001-9931-383X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Junyang Lin","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":null,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2698","last_page":"2711"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12535","display_name":"Machine Learning and Data Classification","score":0.06480000168085098,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.06480000168085098,"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/T10467","display_name":"Psychometric Methodologies and Testing","score":0.04010000079870224,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11875","display_name":"Statistics Education and Methodologies","score":0.03350000083446503,"subfield":{"id":"https://openalex.org/subfields/2613","display_name":"Statistics and Probability"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.5238000154495239},{"id":"https://openalex.org/keywords/feature-selection","display_name":"Feature selection","score":0.3824000060558319},{"id":"https://openalex.org/keywords/bayesian-probability","display_name":"Bayesian probability","score":0.26030001044273376},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.2451000064611435}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5746999979019165},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.5238000154495239},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42250001430511475},{"id":"https://openalex.org/C148483581","wikidata":"https://www.wikidata.org/wiki/Q446488","display_name":"Feature selection","level":2,"score":0.3824000060558319},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3625999987125397},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2946000099182129},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2831000089645386},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.26030001044273376},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.25130000710487366},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.2451000064611435}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.146","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.146","pdf_url":"https://aclanthology.org/2025.findings-emnlp.146.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: EMNLP 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.146","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.146","pdf_url":"https://aclanthology.org/2025.findings-emnlp.146.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: EMNLP 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G6592930967","display_name":null,"funder_award_id":"2023YFF0905400","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"},{"id":"https://openalex.org/G8336371671","display_name":null,"funder_award_id":"U2341229","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320322174","display_name":"People's Government of Jilin Province","ror":"https://ror.org/02fzqav45"},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null},{"id":"https://openalex.org/F4320335787","display_name":"Fundamental Research Funds for the Central Universities","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416035185.pdf","grobid_xml":"https://content.openalex.org/works/W4416035185.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Supervised":[0],"fine-tuning":[1,35],"(SFT)":[2],"is":[3,18,120,189],"crucial":[4],"for":[5,59,150,167,180],"aligning":[6],"Large":[7],"Language":[8],"Models":[9],"(LLMs)":[10],"with":[11,36,105],"human":[12],"instructions.The":[13],"primary":[14],"goal":[15],"during":[16,114],"SFT":[17,71],"to":[19,42,65,98],"select":[20],"a":[21,162],"small":[22],"yet":[23],"representative":[24],"subset":[25,38],"of":[26,69,133],"training":[27,173],"data":[28,54,61,108,118,157],"from":[29],"the":[30,49,67,131],"larger":[31],"pool,":[32],"such":[33,106,151,185],"that":[34,80,93,155],"this":[37,73],"achieves":[39],"results":[40],"comparable":[41],"or":[43],"even":[44],"exceeding":[45],"those":[46],"obtained":[47],"using":[48],"entire":[50],"dataset.However,":[51],"most":[52],"existing":[53],"selection":[55,102,119],"techniques":[56],"are":[57,148],"designed":[58],"small-scale":[60],"pools,":[62],"which":[63],"fail":[64],"meet":[66],"demands":[68],"real-world":[70],"scenarios.In":[72],"paper,":[74],"we":[75,153],"replicated":[76],"several":[77,134],"self-scoring":[78],"methods-those":[79],"do":[81],"not":[82],"rely":[83],"on":[84,126,142,174],"external":[85],"model":[86],"assistance-on":[87],"two":[88],"millionscale":[89],"datasets,":[90],"and":[91,145,164],"found":[92,154],"nearly":[94],"all":[95],"methods":[96],"struggled":[97],"significantly":[99],"outperform":[100],"random":[101],"when":[103,172],"dealing":[104],"largescale":[107,143],"pools.Moreover,":[109],"our":[110],"comparisons":[111],"suggest":[112],"that,":[113],"SFT,":[115],"diversity":[116],"in":[117],"more":[121],"critical":[122],"than":[123],"simply":[124],"focusing":[125],"high-quality":[127],"data.We":[128],"also":[129],"analyzed":[130],"limitations":[132],"current":[135],"approaches,":[136],"explaining":[137],"why":[138,146],"they":[139,147],"perform":[140],"poorly":[141],"datasets":[144],"unsuitable":[149],"contexts.Finally,":[152],"filtering":[156],"by":[158],"token":[159],"length":[160],"offers":[161],"stable":[163],"efficient":[165],"method":[166],"improving":[168],"results.This":[169],"approach,":[170],"particularly":[171],"longtext":[175],"data,":[176],"proves":[177],"highly":[178],"beneficial":[179],"relatively":[181],"weaker":[182],"base":[183],"models,":[184],"as":[186],"Llama3.The":[187],"code":[188],"available":[190],"at":[191],"https://github.com/":[192],"xiatingyu/SFT-DataSelection-at-scale.":[193]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-11-08T00:00:00"}