{"id":"https://openalex.org/W4321011670","doi":"https://doi.org/10.48550/arxiv.2302.06692","title":"Guiding Pretraining in Reinforcement Learning with Large Language Models","display_name":"Guiding Pretraining in Reinforcement Learning with Large Language Models","publication_year":2023,"publication_date":"2023-02-13","ids":{"openalex":"https://openalex.org/W4321011670","doi":"https://doi.org/10.48550/arxiv.2302.06692"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2302.06692","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2302.06692","pdf_url":"https://arxiv.org/pdf/2302.06692","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"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2302.06692","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100352494","display_name":"Yuqing Du","orcid":"https://orcid.org/0000-0002-9147-656X"},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Du, Yuqing","raw_affiliation_strings":["UC Berkeley - University of California [Berkeley] (Berkeley, CA - United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley - University of California [Berkeley] (Berkeley, CA - United States)","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058976886","display_name":"Olivia Watkins","orcid":"https://orcid.org/0000-0003-3054-0385"},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Watkins, Olivia","raw_affiliation_strings":["UC Berkeley - University of California [Berkeley] (Berkeley, CA - United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley - University of California [Berkeley] (Berkeley, CA - United States)","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100380102","display_name":"Zihan Wang","orcid":"https://orcid.org/0009-0006-1175-6163"},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wang, Zihan","raw_affiliation_strings":["University of Washington [Seattle] (Seattle, Washington 98105 - United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Washington [Seattle] (Seattle, Washington 98105 - United States)","institution_ids":["https://openalex.org/I201448701"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048471111","display_name":"C\u00e9dric Colas","orcid":"https://orcid.org/0000-0003-0212-427X"},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]},{"id":"https://openalex.org/I4210102556","display_name":"K Lab (United States)","ror":"https://ror.org/010zs2155","country_code":"US","type":"company","lineage":["https://openalex.org/I4210102556"]},{"id":"https://openalex.org/I4405258931","display_name":"MIT Computer Science and Artificial Intelligence Laboratory","ror":"https://ror.org/01e85r662","country_code":"US","type":"facility","lineage":["https://openalex.org/I4405258931","https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Colas, C\u00e9dric","raw_affiliation_strings":["Flowers - Flowing Epigenetic Robots and Systems (200, avenue de la Vieille Tour \r\n33405 Talence cedex - France)","MIT CSAIL - MIT Computer Science & Artificial Intelligence Lab (United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Flowers - Flowing Epigenetic Robots and Systems (200, avenue de la Vieille Tour \r\n33405 Talence cedex - France)","institution_ids":[]},{"raw_affiliation_string":"MIT CSAIL - MIT Computer Science & Artificial Intelligence Lab (United States)","institution_ids":["https://openalex.org/I1343180700","https://openalex.org/I4210102556","https://openalex.org/I4405258931"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029105520","display_name":"Trevor Darrell","orcid":"https://orcid.org/0000-0001-5453-8533"},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Darrell, Trevor","raw_affiliation_strings":["UC Berkeley - University of California [Berkeley] (Berkeley, CA - United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley - University of California [Berkeley] (Berkeley, CA - United States)","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049349154","display_name":"Pieter Abbeel","orcid":null},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Abbeel, Pieter","raw_affiliation_strings":["UC Berkeley - University of California [Berkeley] (Berkeley, CA - United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley - University of California [Berkeley] (Berkeley, CA - United States)","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100727391","display_name":"Abhishek Gupta","orcid":"https://orcid.org/0000-0002-4282-7407"},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gupta, Abhishek","raw_affiliation_strings":["University of Washington [Seattle] (Seattle, Washington 98105 - United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Washington [Seattle] (Seattle, Washington 98105 - United States)","institution_ids":["https://openalex.org/I201448701"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5070829652","display_name":"Jacob Andreas","orcid":"https://orcid.org/0000-0002-3141-5845"},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]},{"id":"https://openalex.org/I4210102556","display_name":"K Lab (United States)","ror":"https://ror.org/010zs2155","country_code":"US","type":"company","lineage":["https://openalex.org/I4210102556"]},{"id":"https://openalex.org/I4405258931","display_name":"MIT Computer Science and Artificial Intelligence Laboratory","ror":"https://ror.org/01e85r662","country_code":"US","type":"facility","lineage":["https://openalex.org/I4405258931","https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Andreas, Jacob","raw_affiliation_strings":["MIT CSAIL - MIT Computer Science & Artificial Intelligence Lab (United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MIT CSAIL - MIT Computer Science & Artificial Intelligence Lab (United States)","institution_ids":["https://openalex.org/I1343180700","https://openalex.org/I4210102556","https://openalex.org/I4405258931"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":39,"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/T10028","display_name":"Topic Modeling","score":0.9883999824523926,"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":0.9883999824523926,"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":0.9208999872207642,"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/novelty","display_name":"Novelty","score":0.874908447265625},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.7990236282348633},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7231959104537964},{"id":"https://openalex.org/keywords/function","display_name":"Function (biology)","score":0.5083169341087341},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.49472615122795105},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4924558103084564},{"id":"https://openalex.org/keywords/downstream","display_name":"Downstream (manufacturing)","score":0.48264867067337036},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.4398563802242279},{"id":"https://openalex.org/keywords/reinforcement","display_name":"Reinforcement","score":0.4208322763442993},{"id":"https://openalex.org/keywords/human\u2013computer-interaction","display_name":"Human\u2013computer interaction","score":0.3461868166923523},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.2056603729724884},{"id":"https://openalex.org/keywords/social-psychology","display_name":"Social psychology","score":0.10790109634399414},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.08213376998901367},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.06953164935112}],"concepts":[{"id":"https://openalex.org/C2778738651","wikidata":"https://www.wikidata.org/wiki/Q16546687","display_name":"Novelty","level":2,"score":0.874908447265625},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7990236282348633},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7231959104537964},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.5083169341087341},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.49472615122795105},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4924558103084564},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.48264867067337036},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.4398563802242279},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.4208322763442993},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3461868166923523},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.2056603729724884},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.10790109634399414},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.08213376998901367},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.06953164935112},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C78458016","wikidata":"https://www.wikidata.org/wiki/Q840400","display_name":"Evolutionary biology","level":1,"score":0.0},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:arXiv.org:2302.06692","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2302.06692","pdf_url":"https://arxiv.org/pdf/2302.06692","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":"pmh:oai:HAL:hal-04810191v1","is_oa":true,"landing_page_url":"https://inria.hal.science/hal-04810191","pdf_url":"https://inria.hal.science/hal-04810191/document","source":{"id":"https://openalex.org/S4306402512","display_name":"HAL (Le Centre pour la Communication Scientifique Directe)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1294671590","host_organization_name":"Centre National de la Recherche Scientifique","host_organization_lineage":["https://openalex.org/I1294671590"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"ICML 2023 - the 40 th International Conference on Machine Learning,, Jul 2023, Honolulu, United States. ⟨10.48550/arXiv.2302.06692⟩","raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"doi:10.48550/arxiv.2302.06692","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2302.06692","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2302.06692","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2302.06692","pdf_url":"https://arxiv.org/pdf/2302.06692","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"},"sustainable_development_goals":[{"score":0.46000000834465027,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[{"id":"https://openalex.org/G434337789","display_name":null,"funder_award_id":"HORIZON2020","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G4508429655","display_name":"Collaborative Research: RI: Medium: Bootstrapping natural feedback for reinforcement learning","funder_award_id":"2212310","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2381242807","https://openalex.org/W3126131230","https://openalex.org/W2347541121","https://openalex.org/W2080951048","https://openalex.org/W4288804799","https://openalex.org/W3032237421","https://openalex.org/W2390346111","https://openalex.org/W3011883280","https://openalex.org/W2369082698","https://openalex.org/W2401808953"],"abstract_inverted_index":{"Reinforcement":[0],"learning":[1],"algorithms":[2],"typically":[3],"struggle":[4],"in":[5,36,108,114],"the":[6,85,109,115,120],"absence":[7],"of":[8,84,131,145],"a":[9,50,77,82,106,143],"dense,":[10],"well-shaped":[11],"reward":[12],"function.":[13],"Intrinsically":[14],"motivated":[15],"exploration":[16],"methods":[17,32],"address":[18],"this":[19],"limitation":[20],"by":[21,76],"rewarding":[22],"agents":[23,97,127],"for":[24,45,72],"visiting":[25],"novel":[26],"states":[27],"or":[28,139],"transitions,":[29],"but":[30],"these":[31],"offer":[33],"limited":[34],"benefits":[35],"large":[37],"environments":[38],"where":[39],"most":[40],"discovered":[41],"novelty":[42],"is":[43],"irrelevant":[44],"downstream":[46,146],"tasks.":[47,147],"We":[48,111],"describe":[49],"method":[51],"that":[52,125],"uses":[53],"background":[54],"knowledge":[55],"from":[56],"text":[57],"corpora":[58],"to":[59],"shape":[60],"exploration.":[61],"This":[62],"method,":[63],"called":[64],"ELLM":[65,95,113],"(Exploring":[66],"with":[67,81],"LLMs)":[68],"rewards":[69],"an":[70],"agent":[71],"achieving":[73],"goals":[74],"suggested":[75],"language":[78,92],"model":[79,93],"prompted":[80],"description":[83],"agent's":[86],"current":[87],"state.":[88],"By":[89],"leveraging":[90],"large-scale":[91],"pretraining,":[94],"guides":[96],"toward":[98],"human-meaningful":[99],"and":[100,119,136],"plausibly":[101],"useful":[102],"behaviors":[103,133],"without":[104],"requiring":[105],"human":[107],"loop.":[110],"evaluate":[112],"Crafter":[116],"game":[117],"environment":[118],"Housekeep":[121],"robotic":[122],"simulator,":[123],"showing":[124],"ELLM-trained":[126],"have":[128],"better":[129],"coverage":[130],"common-sense":[132],"during":[134],"pretraining":[135],"usually":[137],"match":[138],"improve":[140],"performance":[141],"on":[142],"range":[144],"Code":[148],"available":[149],"at":[150],"https://github.com/yuqingd/ellm.":[151]},"counts_by_year":[{"year":2025,"cited_by_count":12},{"year":2024,"cited_by_count":19},{"year":2023,"cited_by_count":8}],"updated_date":"2026-08-21T09:56:20.448147","created_date":"2025-10-10T00:00:00"}