{"id":"https://openalex.org/W4410105421","doi":"https://doi.org/10.52202/085713-2158","title":"Self-Generated In-Context Examples Improve LLM Agents for Sequential Decision-Making Tasks","display_name":"Self-Generated In-Context Examples Improve LLM Agents for Sequential Decision-Making Tasks","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4410105421","doi":"https://doi.org/10.52202/085713-2158"},"language":"en","primary_location":{"id":"doi:10.52202/085713-2158","doi":"https://doi.org/10.52202/085713-2158","is_oa":false,"landing_page_url":"https://doi.org/10.52202/085713-2158","pdf_url":null,"source":{"id":"https://openalex.org/S4363606243","display_name":"neural information processing systems","issn_l":"1049-5258","issn":["1049-5258"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","norway-2"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Advances in Neural Information Processing Systems 38","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2505.00234","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5069897950","display_name":"Vishnu Sarukkai","orcid":"https://orcid.org/0009-0006-9809-9994","observed_orcids":["https://orcid.org/0009-0006-9809-9994"]},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Vishnu Sarukkai","raw_affiliation_strings":["Stanford"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford","institution_ids":["https://openalex.org/I97018004"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101504251","display_name":"Zhiqiang Xie","orcid":"https://orcid.org/0000-0001-9828-3034","observed_orcids":["https://orcid.org/0000-0001-9828-3034"]},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhiqiang Xie","raw_affiliation_strings":["Stanford University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford University","institution_ids":["https://openalex.org/I97018004"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5037444018","display_name":"Kayvon Fatahalian","orcid":"https://orcid.org/0000-0001-8754-0429","observed_orcids":["https://orcid.org/0000-0001-8754-0429"]},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kayvon Fatahalian","raw_affiliation_strings":["Stanford"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford","institution_ids":["https://openalex.org/I97018004"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I97018004"],"apc_list":null,"apc_paid":null,"fwci":7.4902,"has_fulltext":true,"cited_by_count":21,"citation_normalized_percentile":{"value":0.97768924,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"71753","last_page":"71786"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10906","display_name":"AI-based Problem Solving and Planning","score":0.7099999785423279,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.7099999785423279,"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":0.06560000032186508,"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":0.04320000112056732,"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/llm-based-agents","display_name":"LLM agents","score":0.8199999928474426},{"id":"https://openalex.org/keywords/sequential-decision-making","display_name":"sequential decision making","score":0.9330000281333923},{"id":"https://openalex.org/keywords/in-context-learning","display_name":"in-context learning","score":0.8389999866485596},{"id":"https://openalex.org/keywords/population-based-training","display_name":"population-based training","score":0.9660000205039978},{"id":"https://openalex.org/keywords/exemplar-selection","display_name":"exemplar selection","score":0.7889999747276306},{"id":"https://openalex.org/keywords/alfworld","display_name":"ALFWorld","score":0.8519999980926514}],"concepts":[{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6034618020057678},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5764216184616089},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35548949241638184},{"id":"https://openalex.org/C539667460","wikidata":"https://www.wikidata.org/wiki/Q2414942","display_name":"Management science","level":1,"score":0.34370261430740356},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.1305696666240692},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.05326327681541443},{"id":"https://openalex.org/C166957645","wikidata":"https://www.wikidata.org/wiki/Q23498","display_name":"Archaeology","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.52202/085713-2158","doi":"https://doi.org/10.52202/085713-2158","is_oa":false,"landing_page_url":"https://doi.org/10.52202/085713-2158","pdf_url":null,"source":{"id":"https://openalex.org/S4363606243","display_name":"neural information processing systems","issn_l":"1049-5258","issn":["1049-5258"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","norway-2"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Advances in Neural Information Processing Systems 38","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2505.00234","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2505.00234","pdf_url":"https://arxiv.org/pdf/2505.00234","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2505.00234","doi":"https://doi.org/10.48550/arxiv.2505.00234","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2505.00234","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,"listed_in":[],"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":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2505.00234","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2505.00234","pdf_url":"https://arxiv.org/pdf/2505.00234","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4410105421.pdf","grobid_xml":"https://content.openalex.org/works/W4410105421.grobid-xml"},"referenced_works_count":27,"referenced_works":["https://openalex.org/W2952862139","https://openalex.org/W3043369204","https://openalex.org/W3126325318","https://openalex.org/W3134642945","https://openalex.org/W4223908421","https://openalex.org/W4310509152","https://openalex.org/W4323697341","https://openalex.org/W4378505261","https://openalex.org/W4387430850","https://openalex.org/W4387561453","https://openalex.org/W4391631574","https://openalex.org/W4394662461","https://openalex.org/W4402500863","https://openalex.org/W4402671629","https://openalex.org/W4402733601","https://openalex.org/W4403524626","https://openalex.org/W4403571374","https://openalex.org/W4403580146","https://openalex.org/W4403853635","https://openalex.org/W4404782958","https://openalex.org/W4405766390","https://openalex.org/W4411119282","https://openalex.org/W4414281281","https://openalex.org/W4415124086","https://openalex.org/W4415795430","https://openalex.org/W4415799296","https://openalex.org/W7160101477"],"related_works":["https://openalex.org/W4400525461","https://openalex.org/W7135059314","https://openalex.org/W4404781773","https://openalex.org/W7133187879","https://openalex.org/W7084156219","https://openalex.org/W7162239154","https://openalex.org/W4415108122","https://openalex.org/W4412945322","https://openalex.org/W7140993884","https://openalex.org/W4412394791"],"abstract_inverted_index":{"Improving":[0],"Large":[1],"Language":[2],"Model":[3],"(LLM)":[4],"agents":[5,29,166],"for":[6,57],"sequential":[7],"decision-making":[8],"tasks":[9],"typically":[10],"requires":[11],"extensive":[12],"task-specific":[13],"knowledge":[14,178],"engineering--custom":[15],"prompts,":[16],"curated":[17],"examples,":[18],"and":[19,45,82,98,126,157],"specialized":[20],"observation/action":[21],"spaces.":[22],"We":[23,108],"investigate":[24],"a":[25,47,173],"different":[26],"approach":[27,112],"where":[28],"automatically":[30],"improve":[31,169],"by":[32,92],"learning":[33],"from":[34,94],"their":[35,135],"own":[36],"successful":[37,64],"experiences":[38],"without":[39],"human":[40],"intervention.":[41],"Our":[42,160],"method":[43,145],"constructs":[44],"refines":[46],"database":[48],"of":[49,63,102],"self-generated":[50],"trajectories":[51,65,132],"that":[52,129,152,165],"serve":[53],"as":[54,138],"in-context":[55,139],"examples":[56],"future":[58],"tasks.":[59],"Even":[60],"naive":[61],"accumulation":[62],"yields":[66],"substantial":[67],"performance":[68,101],"gains":[69],"across":[70],"three":[71],"diverse":[72],"benchmarks:":[73],"ALFWorld":[74],"(73%":[75],"to":[76,80,85,96,121,176],"89%),":[77],"Wordcraft":[78],"(55%":[79],"64%),":[81],"InterCode-SQL":[83],"(75%":[84],"79%).":[86],"These":[87],"improvements":[88],"exceed":[89],"those":[90],"achieved":[91],"upgrading":[93],"gpt-4o-mini":[95],"gpt-4o":[97],"match":[99],"the":[100],"allowing":[103],"multiple":[104],"attempts":[105],"per":[106],"task.":[107],"further":[109],"enhance":[110],"this":[111],"with":[113],"two":[114],"innovations:":[115],"database-level":[116],"curation":[117,128],"using":[118],"population-based":[119],"training":[120],"propagate":[122],"high-performing":[123],"example":[124],"collections,":[125],"exemplar-level":[127],"selectively":[130],"retains":[131],"based":[133],"on":[134,149],"empirical":[136],"utility":[137],"examples.":[140],"With":[141],"these":[142],"enhancements,":[143],"our":[144],"achieves":[146],"93%":[147],"success":[148],"ALFWorld--surpassing":[150],"approaches":[151],"use":[153],"more":[154],"powerful":[155],"LLMs":[156],"hand-crafted":[158],"components.":[159],"trajectory":[161],"bootstrapping":[162],"technique":[163],"demonstrates":[164],"can":[167],"autonomously":[168],"through":[170],"experience,":[171],"offering":[172],"scalable":[174],"alternative":[175],"labor-intensive":[177],"engineering.":[179]},"counts_by_year":[{"year":2026,"cited_by_count":17},{"year":2025,"cited_by_count":4}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2025-05-06T00:00:00"}