{"id":"https://openalex.org/W7138842930","doi":"https://doi.org/10.48550/arxiv.2603.17378","title":"Efficient Exploration at Scale","display_name":"Efficient Exploration at Scale","publication_year":2026,"publication_date":"2026-03-18","ids":{"openalex":"https://openalex.org/W7138842930","doi":"https://doi.org/10.48550/arxiv.2603.17378"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.17378","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17378","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.2603.17378","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5083140383","display_name":"Seyed Mohammad Asghari","orcid":"https://orcid.org/0000-0003-1614-4669"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Asghari, Seyed Mohammad","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071576173","display_name":"Chris Chute","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chute, Chris","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085692306","display_name":"Vikranth Dwaracherla","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dwaracherla, Vikranth","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129792981","display_name":"Xiuyuan Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Xiuyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080200687","display_name":"Mehdi Jafarnia","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jafarnia, Mehdi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018787411","display_name":"Victor Minden","orcid":"https://orcid.org/0009-0006-9084-6289"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Minden, Victor","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129972399","display_name":"Zheng Wen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wen, Zheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130116184","display_name":"Benjamin Van Roy","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Van Roy, Benjamin","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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.1527000069618225,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.1527000069618225,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.1266999989748001,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.11840000003576279,"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.7750999927520752},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.5223000049591064},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.49470001459121704},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.47760000824928284},{"id":"https://openalex.org/keywords/variation","display_name":"Variation (astronomy)","score":0.4602999985218048},{"id":"https://openalex.org/keywords/reinforcement","display_name":"Reinforcement","score":0.40709999203681946},{"id":"https://openalex.org/keywords/data-modeling","display_name":"Data modeling","score":0.32420000433921814}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7750999927520752},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7423999905586243},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6328999996185303},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5893999934196472},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.5223000049591064},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.49470001459121704},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.47760000824928284},{"id":"https://openalex.org/C2778334786","wikidata":"https://www.wikidata.org/wiki/Q1586270","display_name":"Variation (astronomy)","level":2,"score":0.4602999985218048},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.40709999203681946},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.32420000433921814},{"id":"https://openalex.org/C196340769","wikidata":"https://www.wikidata.org/wiki/Q7698910","display_name":"Temporal difference learning","level":3,"score":0.3222000002861023},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.3192000091075897},{"id":"https://openalex.org/C2780102126","wikidata":"https://www.wikidata.org/wiki/Q10928179","display_name":"Online and offline","level":2,"score":0.3176000118255615},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.29420000314712524},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.2919999957084656},{"id":"https://openalex.org/C74672266","wikidata":"https://www.wikidata.org/wiki/Q815859","display_name":"Language acquisition","level":2,"score":0.2858999967575073},{"id":"https://openalex.org/C2986087404","wikidata":"https://www.wikidata.org/wiki/Q15946010","display_name":"Online learning","level":2,"score":0.2858999967575073},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.26570001244544983},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2524999976158142}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.17378","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17378","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.2603.17378","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17378","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":{"We":[0],"develop":[1],"an":[2,75],"online":[3],"learning":[4,14],"algorithm":[5,20,93,125],"that":[6,79,153],"dramatically":[7],"improves":[8],"the":[9,38,42,57,63,95,148],"data":[10,29,116],"efficiency":[11,64],"of":[12,50,97],"reinforcement":[13,53,73],"from":[15,119],"human":[16],"feedback":[17],"(RLHF).":[18],"Our":[19],"incrementally":[21],"updates":[22],"reward":[23,33,58,81],"and":[24,83],"language":[25,43,89],"models":[26,80,90],"as":[27],"choice":[28,39],"is":[30,35,45],"received.":[31],"The":[32],"model":[34,44],"fit":[36],"to":[37,71,130,151],"data,":[40],"while":[41],"updated":[46],"by":[47,56],"a":[48,66,112,140],"variation":[49],"reinforce,":[51],"with":[52],"signals":[54],"provided":[55],"model.":[59],"Several":[60],"features":[61],"enable":[62],"gains:":[65],"small":[67],"affirmative":[68],"nudge":[69],"added":[70],"each":[72],"signal,":[74],"epistemic":[76],"neural":[77],"network":[78],"uncertainty,":[82],"information-directed":[84],"exploration.":[85],"With":[86],"Gemma":[87],"large":[88,155],"(LLMs),":[91],"our":[92,120,124,144],"matches":[94],"performance":[96],"offline":[98,132],"RLHF":[99,133],"trained":[100,126,134],"on":[101,127,135],"200K":[102],"labels":[103,129],"using":[104],"fewer":[105],"than":[106,111],"20K":[107],"labels,":[108],"representing":[109],"more":[110],"10x":[113],"gain":[114],"in":[115],"efficiency.":[117],"Extrapolating":[118],"results,":[121],"we":[122],"expect":[123],"1M":[128],"match":[131],"1B":[136],"labels.":[137],"This":[138],"represents":[139],"1,000x":[141],"gain.":[142],"To":[143],"knowledge,":[145],"these":[146],"are":[147,157],"first":[149],"results":[150],"demonstrate":[152],"such":[154],"improvements":[156],"possible.":[158]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-20T00:00:00"}