{"id":"https://openalex.org/W4412394791","doi":"https://doi.org/10.1145/3726302.3729965","title":"Empowering Large Language Model Agent through Step-Level Self-Critique and Self-Training","display_name":"Empowering Large Language Model Agent through Step-Level Self-Critique and Self-Training","publication_year":2025,"publication_date":"2025-07-13","ids":{"openalex":"https://openalex.org/W4412394791","doi":"https://doi.org/10.1145/3726302.3729965"},"language":"en","primary_location":{"id":"doi:10.1145/3726302.3729965","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3726302.3729965","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3726302.3729965","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3726302.3729965","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5073132517","display_name":"Yuanzhao Zhai","orcid":"https://orcid.org/0000-0003-1385-0074"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuanzhao Zhai","raw_affiliation_strings":["National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China"],"raw_orcid":"https://orcid.org/0000-0003-1385-0074","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Huanxi Liu","orcid":"https://orcid.org/0009-0001-6526-3333"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huanxi Liu","raw_affiliation_strings":["National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China"],"raw_orcid":"https://orcid.org/0009-0001-6526-3333","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108148866","display_name":"Zhuo Zhang","orcid":"https://orcid.org/0000-0002-1835-5411"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]},{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhuo Zhang","raw_affiliation_strings":["Harbin Institute of Technology (Shenzhen), Shenzhen, China and Peng Cheng Laboratory, Shenzhen, China"],"raw_orcid":"https://orcid.org/0000-0002-1835-5411","affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology (Shenzhen), Shenzhen, China and Peng Cheng Laboratory, Shenzhen, China","institution_ids":["https://openalex.org/I204983213","https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021926421","display_name":"Tong Lin","orcid":null},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tong Lin","raw_affiliation_strings":["National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China"],"raw_orcid":"https://orcid.org/0009-0008-3907-9402","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013340793","display_name":"Kele Xu","orcid":"https://orcid.org/0000-0001-5997-5169"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Kele Xu","raw_affiliation_strings":["National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China"],"raw_orcid":"https://orcid.org/0000-0001-5997-5169","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078272955","display_name":"Cheng Yang","orcid":"https://orcid.org/0000-0002-4782-1645"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Cheng Yang","raw_affiliation_strings":["National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China"],"raw_orcid":"https://orcid.org/0000-0002-4782-1645","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039795290","display_name":"Dawei Feng","orcid":"https://orcid.org/0000-0002-7587-8905"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dawei Feng","raw_affiliation_strings":["National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China"],"raw_orcid":"https://orcid.org/0000-0002-7587-8905","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088885490","display_name":"Bo Ding","orcid":"https://orcid.org/0000-0002-1236-8318"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bo Ding","raw_affiliation_strings":["National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China"],"raw_orcid":"https://orcid.org/0000-0002-1236-8318","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101522100","display_name":"Huaimin Wang","orcid":"https://orcid.org/0000-0002-3245-1901"},"institutions":[{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huaimin Wang","raw_affiliation_strings":["National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China"],"raw_orcid":"https://orcid.org/0000-0002-3245-1901","affiliations":[{"raw_affiliation_string":"National University of Defense Technology, Changsha, China and State Key Laboratory of Complex & Critical Software Environment, Changsha, China","institution_ids":["https://openalex.org/I170215575"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1444","last_page":"1454"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11574","display_name":"Artificial Intelligence in Games","score":0.9994000196456909,"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/T11574","display_name":"Artificial Intelligence in Games","score":0.9994000196456909,"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":0.9972000122070312,"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/T11975","display_name":"Evolutionary Algorithms and Applications","score":0.982699990272522,"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/computer-science","display_name":"Computer science","score":0.677922248840332},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.5850502252578735},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.35245925188064575}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.677922248840332},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5850502252578735},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35245925188064575},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3726302.3729965","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3726302.3729965","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3726302.3729965","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3726302.3729965","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3726302.3729965","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3726302.3729965","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G8246084123","display_name":null,"funder_award_id":"2023ZD0121101","funder_id":"https://openalex.org/F4320323817","funder_display_name":"Universitas Brawijaya"}],"funders":[{"id":"https://openalex.org/F4320323817","display_name":"Universitas Brawijaya","ror":"https://ror.org/01wk3d929"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412394791.pdf","grobid_xml":"https://content.openalex.org/works/W4412394791.grobid-xml"},"referenced_works_count":10,"referenced_works":["https://openalex.org/W1593271688","https://openalex.org/W1625390266","https://openalex.org/W2126316555","https://openalex.org/W2990138404","https://openalex.org/W4226278401","https://openalex.org/W4283828996","https://openalex.org/W4389520747","https://openalex.org/W4393065402","https://openalex.org/W4400525725","https://openalex.org/W4409362104"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W230091440","https://openalex.org/W2390279801","https://openalex.org/W2233261550","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2810751659"],"abstract_inverted_index":{"Large":[0],"Language":[1],"Model":[2],"(LLM)":[3],"agents":[4,64,102,128,177,204],"frequently":[5],"produce":[6],"sub-optimal":[7],"actions":[8,91],"when":[9],"tackling":[10],"complex,":[11],"multi-step":[12],"decision-making":[13],"tasks.":[14],"Employing":[15],"self-critique":[16,93],"to":[17,39,61,141],"identify":[18,40],"flaws":[19],"and":[20,68,87,94,155,163,183,196],"suggest":[21],"enhancements":[22],"is":[23,33,153],"an":[24,82],"effective":[25],"strategy":[26],"for":[27],"refining":[28,164],"actions.":[29],"Although":[30],"trajectory-level":[31],"critique":[32],"commonly":[34],"employed,":[35],"it":[36],"often":[37],"fails":[38],"flawed":[41],"steps":[42],"accurately.":[43],"In":[44],"this":[45],"paper,":[46],"we":[47],"introduce":[48],"SLSC-MCTS,":[49,76,100],"a":[50,130],"method":[51,124],"that":[52,172],"integrates":[53],"Monte":[54],"Carlo":[55],"Tree":[56],"Search":[57],"with":[58,75],"Step-Level":[59],"Self-Critique":[60],"enhance":[62],"LLM":[63,78,101,127,145,167,176,203],"during":[65,178],"both":[66],"testing":[67],"self-training":[69,131,148],"phases.":[70],"During":[71],"decision":[72,139],"tree":[73,140],"expansion":[74],"the":[77,118,137,144,160],"agent":[79],"initially":[80],"generates":[81,89],"action,":[83],"receives":[84],"environmental":[85],"feedback,":[86],"subsequently":[88],"further":[90,125,158],"via":[92,151],"refinement.":[95],"Through":[96],"multiple":[97],"episodes":[98],"of":[99,166,201],"can":[103],"effectively":[104],"utilize":[105],"step-level":[106],"critiques":[107,119],"while":[108],"disregarding":[109],"ineffective":[110],"ones":[111],"based":[112],"on":[113],"node":[114],"values,":[115],"thereby":[116],"incorporating":[117],"more":[120],"robustly.":[121],"Additionally,":[122],"our":[123],"empowers":[126],"in":[129,215],"manner,":[132],"collecting":[133],"training":[134],"data":[135,149],"from":[136],"constructed":[138],"iteratively":[142],"fine-tune":[143],"agents.":[146,168],"The":[147],"gathered":[150],"SLSC-MCTS":[152,173],"diverse":[154],"high-quality,":[156],"which":[157],"enhances":[159],"reasoning,":[161],"critiquing,":[162],"abilities":[165],"Experimental":[169],"results":[170],"demonstrate":[171],"significantly":[174],"improves":[175],"testing,":[179],"surpassing":[180,212],"state-of-the-art":[181],"baselines":[182],"achieving":[184],"shorter":[185],"task":[186],"completion":[187],"trajectories":[188],"across":[189],"information":[190],"retrieval":[191],"benchmarks":[192],"such":[193],"as":[194],"WebShop":[195],"HotPotQA.":[197],"After":[198],"three":[199],"iterations":[200],"self-training,":[202],"established":[205],"by":[206],"Llama-3.1-8B-Instruct":[207],"show":[208],"substantial":[209],"improvement,":[210],"even":[211],"human":[213],"experts":[214],"WebShop.":[216]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-22T07:51:19.307946","created_date":"2025-10-10T00:00:00"}