{"id":"https://openalex.org/W4412888257","doi":"https://doi.org/10.18653/v1/2025.findings-acl.604","title":"STeCa: Step-level Trajectory Calibration for LLM Agent Learning","display_name":"STeCa: Step-level Trajectory Calibration for LLM Agent Learning","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412888257","doi":"https://doi.org/10.18653/v1/2025.findings-acl.604"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.604","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.604","pdf_url":"https://aclanthology.org/2025.findings-acl.604.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: ACL 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-acl.604.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100637385","display_name":"Hanlin Wang","orcid":"https://orcid.org/0000-0002-5416-9392"},"institutions":[{"id":"https://openalex.org/I14243506","display_name":"Hong Kong Polytechnic University","ror":"https://ror.org/0030zas98","country_code":"HK","type":"education","lineage":["https://openalex.org/I14243506"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Hanlin Wang","raw_affiliation_strings":["Department of Computing , The Hong Kong Polytechnic University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computing , The Hong Kong Polytechnic University","institution_ids":["https://openalex.org/I14243506"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012676602","display_name":"Jian Wang","orcid":"https://orcid.org/0000-0002-8992-8336"},"institutions":[{"id":"https://openalex.org/I14243506","display_name":"Hong Kong Polytechnic University","ror":"https://ror.org/0030zas98","country_code":"HK","type":"education","lineage":["https://openalex.org/I14243506"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Jian Wang","raw_affiliation_strings":["Department of Computing , The Hong Kong Polytechnic University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computing , The Hong Kong Polytechnic University","institution_ids":["https://openalex.org/I14243506"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5009277283","display_name":"Chak Tou Leong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chak Tou Leong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100408983","display_name":"Wenjie Li","orcid":"https://orcid.org/0000-0002-7360-8864"},"institutions":[{"id":"https://openalex.org/I14243506","display_name":"Hong Kong Polytechnic University","ror":"https://ror.org/0030zas98","country_code":"HK","type":"education","lineage":["https://openalex.org/I14243506"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Wenjie Li","raw_affiliation_strings":["Department of Computing , The Hong Kong Polytechnic University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computing , The Hong Kong Polytechnic University","institution_ids":["https://openalex.org/I14243506"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"11597","last_page":"11614"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T14470","display_name":"Advanced Data Processing Techniques","score":0.9625999927520752,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T14470","display_name":"Advanced Data Processing Techniques","score":0.9625999927520752,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.9350000023841858,"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":0.9301000237464905,"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/trajectory","display_name":"Trajectory","score":0.8406141400337219},{"id":"https://openalex.org/keywords/calibration","display_name":"Calibration","score":0.7221173048019409},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6929131746292114},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.38893282413482666},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.09815815091133118},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.07457166910171509}],"concepts":[{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.8406141400337219},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.7221173048019409},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6929131746292114},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38893282413482666},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.09815815091133118},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.07457166910171509},{"id":"https://openalex.org/C1276947","wikidata":"https://www.wikidata.org/wiki/Q333","display_name":"Astronomy","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.604","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.604","pdf_url":"https://aclanthology.org/2025.findings-acl.604.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: ACL 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.604","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.604","pdf_url":"https://aclanthology.org/2025.findings-acl.604.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: ACL 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412888257.pdf","grobid_xml":"https://content.openalex.org/works/W4412888257.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4323768008"],"abstract_inverted_index":{"Large":[0],"language":[1],"model":[2],"(LLM)-based":[3],"agents":[4,50,106,138],"have":[5],"shown":[6],"promise":[7],"in":[8],"tackling":[9],"complex":[10],"tasks":[11,141],"by":[12,47],"interacting":[13],"dynamically":[14],"with":[15,118,142],"the":[16,61,67],"environment.Existing":[17],"work":[18],"primarily":[19],"focuses":[20],"on":[21],"behavior":[22],"cloning":[23],"from":[24,53,109],"expert":[25],"demonstrations":[26],"or":[27],"preference":[28],"learning":[29],"through":[30,92],"exploratory":[31],"trajectory":[32],"sampling.However,":[33],"these":[34,115],"methods":[35],"often":[36],"struggle":[37],"to":[38,51,69,107,139],"address":[39,57],"long-horizon":[40],"tasks,":[41],"where":[42],"suboptimal":[43,90],"actions":[44,91],"accumulate":[45],"step":[46],"step,":[48],"causing":[49],"deviate":[52],"correct":[54],"task":[55],"trajectories.To":[56],"this,":[58],"we":[59],"highlight":[60],"importance":[62],"of":[63],"timely":[64,135],"calibration":[65,72,136],"and":[66],"need":[68],"automatically":[70],"construct":[71],"trajectories":[73,101,117,120],"for":[74,84,121],"training":[75],"agents.We":[76],"proposeStep-Level":[77],"Trajectory":[78],"Calibration":[79],"(STeCa),":[80],"a":[81,93],"novel":[82],"framework":[83],"LLM":[85],"agent":[86],"learning.Specifically,":[87],"STeCa":[88,127],"identifies":[89],"step-level":[94],"reward":[95],"comparison":[96],"during":[97],"exploration.It":[98],"constructs":[99],"calibrated":[100,116],"using":[102],"LLM-driven":[103],"reflection,":[104],"enabling":[105],"learn":[108],"improved":[110],"decisionmaking":[111],"processes.We":[112],"finally":[113],"leverage":[114],"successful":[119],"reinforced":[122],"training.Extensive":[123],"experiments":[124],"demonstrate":[125],"that":[126,134],"significantly":[128],"outperforms":[129],"existing":[130],"methods.Further":[131],"analysis":[132],"highlights":[133],"enables":[137],"complete":[140],"greater":[143],"robustness.":[144]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}