{"id":"https://openalex.org/W7134960605","doi":"https://doi.org/10.48550/arxiv.2603.09692","title":"ActiveUltraFeedback: Efficient Preference Data Generation using Active Learning","display_name":"ActiveUltraFeedback: Efficient Preference Data Generation using Active Learning","publication_year":2026,"publication_date":"2026-03-10","ids":{"openalex":"https://openalex.org/W7134960605","doi":"https://doi.org/10.48550/arxiv.2603.09692"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.09692","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.09692","pdf_url":null,"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.09692","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5128772734","display_name":"Davit Melikidze","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Melikidze, Davit","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128783703","display_name":"Marian Schneider","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Schneider, Marian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128704773","display_name":"Jessica Lam","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lam, Jessica","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128764471","display_name":"Martin Wertich","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wertich, Martin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128756335","display_name":"Ido Hakimi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hakimi, Ido","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127861793","display_name":"Barna P\u00e1sztor","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"P\u00e1sztor, Barna","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5128758089","display_name":"Andreas Krause","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Krause, Andreas","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":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12072","display_name":"Machine Learning and Algorithms","score":0.2257000058889389,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.2257000058889389,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.12060000002384186,"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/T10028","display_name":"Topic Modeling","score":0.10320000350475311,"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/pipeline","display_name":"Pipeline (software)","score":0.7260000109672546},{"id":"https://openalex.org/keywords/active-learning","display_name":"Active learning (machine learning)","score":0.5963000059127808},{"id":"https://openalex.org/keywords/modular-design","display_name":"Modular design","score":0.5950000286102295},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.5785999894142151},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.5080999732017517},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.454800009727478},{"id":"https://openalex.org/keywords/preference-learning","display_name":"Preference learning","score":0.42179998755455017}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7350999712944031},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.7260000109672546},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6521999835968018},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.597599983215332},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.5963000059127808},{"id":"https://openalex.org/C101468663","wikidata":"https://www.wikidata.org/wiki/Q1620158","display_name":"Modular design","level":2,"score":0.5950000286102295},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.5785999894142151},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.5080999732017517},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.454800009727478},{"id":"https://openalex.org/C181204326","wikidata":"https://www.wikidata.org/wiki/Q7239820","display_name":"Preference learning","level":3,"score":0.42179998755455017},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.4131999909877777},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.4068000018596649},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.37290000915527344},{"id":"https://openalex.org/C24756922","wikidata":"https://www.wikidata.org/wiki/Q1757694","display_name":"Data quality","level":3,"score":0.3070000112056732},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.3027999997138977},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.2775999903678894},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.26460000872612},{"id":"https://openalex.org/C2777868144","wikidata":"https://www.wikidata.org/wiki/Q7239817","display_name":"Preference elicitation","level":3,"score":0.26330000162124634},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.25130000710487366}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.09692","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.09692","pdf_url":null,"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.09692","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.09692","pdf_url":null,"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reinforcement":[0],"Learning":[1],"from":[2],"Human":[3],"Feedback":[4],"(RLHF)":[5],"has":[6],"become":[7],"the":[8,22,53,62,129],"standard":[9,66],"for":[10,57,99],"aligning":[11],"Large":[12],"Language":[13],"Models":[14],"(LLMs),":[15],"yet":[16],"its":[17],"efficacy":[18],"is":[19,138],"bottlenecked":[20],"by":[21],"high":[23],"cost":[24],"of":[25,65,128],"acquiring":[26],"preference":[27,144],"data,":[28],"especially":[29],"in":[30,114],"low-resource":[31],"and":[32,76,142],"expert":[33],"domains.":[34],"To":[35],"address":[36],"this,":[37],"we":[38],"introduce":[39],"ACTIVEULTRAFEEDBACK,":[40],"a":[41],"modular":[42],"active":[43],"learning":[44],"pipeline":[45,60,137],"that":[46,93,104,109],"leverages":[47],"uncertainty":[48],"estimates":[49],"to":[50,111,133],"dynamically":[51],"identify":[52],"most":[54],"informative":[55],"responses":[56],"annotation.":[58],"Our":[59,101,136],"facilitates":[61],"systematic":[63],"evaluation":[64],"response":[67,82],"selection":[68],"methods":[69,80],"alongside":[70],"DOUBLE":[71],"REVERSE":[72],"THOMPSON":[73],"SAMPLING":[74],"(DRTS)":[75],"DELTAUCB,":[77],"two":[78],"novel":[79],"prioritizing":[81],"pairs":[83,95],"with":[84,123],"large":[85],"predicted":[86],"quality":[87],"gaps,":[88],"leveraging":[89],"recent":[90],"results":[91,122],"showing":[92],"such":[94],"provide":[96],"good":[97],"signals":[98],"fine-tuning.":[100],"experiments":[102],"demonstrate":[103],"ACTIVEULTRAFEEDBACK":[105],"yields":[106],"high-quality":[107],"datasets":[108,145],"lead":[110],"significant":[112],"improvements":[113],"downstream":[115],"performance,":[116],"notably":[117],"achieving":[118],"comparable":[119],"or":[120],"superior":[121],"as":[124,126],"little":[125],"one-sixth":[127],"annotated":[130],"data":[131],"relative":[132],"static":[134],"baselines.":[135],"available":[139],"at":[140,146],"https://github.com/lasgroup/ActiveUltraFeedback":[141],"our":[143],"https://huggingface.co/ActiveUltraFeedback.":[147]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-12T00:00:00"}