{"id":"https://openalex.org/W4412945419","doi":"https://doi.org/10.18653/v1/2025.acl-long.559","title":"SoRFT: Issue Resolving with Subtask-oriented Reinforced Fine-Tuning","display_name":"SoRFT: Issue Resolving with Subtask-oriented Reinforced Fine-Tuning","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412945419","doi":"https://doi.org/10.18653/v1/2025.acl-long.559"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.559","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.559","pdf_url":"https://aclanthology.org/2025.acl-long.559.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-long.559.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5119204520","display_name":"Zexiong Ma","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zexiong Ma","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119204521","display_name":"Chao Peng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chao Peng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119204522","display_name":"Pengfei Gao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pengfei Gao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119204523","display_name":"Xiangxin Meng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiangxin Meng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119204524","display_name":"Yanzhen Zou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yanzhen Zou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5119204525","display_name":"Bing Xie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bing Xie","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":true,"cited_by_count":5,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"11427","last_page":"11441"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11338","display_name":"Advancements in Photolithography Techniques","score":0.8679999709129333,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic 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/T11338","display_name":"Advancements in Photolithography Techniques","score":0.8679999709129333,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic 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/T11032","display_name":"VLSI and Analog Circuit Testing","score":0.8259999752044678,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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":0.6460134983062744}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6460134983062744}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.559","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.559","pdf_url":"https://aclanthology.org/2025.acl-long.559.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.acl-long.559","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.559","pdf_url":"https://aclanthology.org/2025.acl-long.559.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G3001381272","display_name":null,"funder_award_id":"2023YFB4503803","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412945419.pdf","grobid_xml":"https://content.openalex.org/works/W4412945419.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/W4396696052"],"abstract_inverted_index":{"Mainstream":[0],"issue-resolving":[1,132],"frameworks":[2],"predominantly":[3],"rely":[4],"on":[5,103,120],"commercial":[6,143],"models,":[7],"leading":[8],"to":[9,26,41,142],"high":[10],"costs":[11],"and":[12,24,61,87,106,137],"privacy":[13],"concerns.Existing":[14],"training":[15,39,68],"approaches":[16],"for":[17],"issue":[18,44,50],"resolving":[19,45,51],"struggle":[20],"with":[21,95,123],"poor":[22],"generalization":[23],"fail":[25],"fully":[27],"leverage":[28],"open-source":[29,114],"development":[30],"resources.We":[31],"propose":[32],"Subtask-oriented":[33],"Reinforced":[34],"Fine-Tuning":[35],"(SoRFT),":[36],"a":[37,139],"novel":[38],"approach":[40],"enhance":[42],"the":[43,85,100],"capability":[46],"of":[47,66,75],"LLMs.We":[48],"decomposes":[49],"into":[52],"structured":[53],"subtasks:":[54],"file":[55],"localization,":[56,58,60],"function":[57],"line":[59],"code":[62],"edit":[63],"generation.SoRFT":[64],"consists":[65],"two":[67],"stages:":[69],"(1)":[70],"rejection-sampled":[71],"supervised":[72],"fine-tuning,":[73],"Chain":[74],"Thought":[76],"(CoT)":[77],"data":[78],"is":[79],"filtered":[80],"using":[81],"ground-truth":[82,96],"before":[83],"fine-tuning":[84],"LLM,":[86],"(2)":[88],"rule-based":[89],"reinforcement":[90],"learning,":[91],"which":[92],"leverages":[93],"PPO":[94],"based":[97],"rewards.We":[98],"evaluate":[99],"SoRFT-trained":[101],"model":[102,135],"SWE-Bench":[104,107,121],"Verified":[105,122],"Lite,":[108],"achieving":[109],"state-of-the-art":[110],"(SOTA)":[111],"performance":[112],"among":[113],"models":[115],"(e.g.,":[116],"resolve":[117],"21.4%":[118],"issues":[119],"SoRFT-Qwen-7B).The":[124],"experimental":[125],"results":[126],"demonstrate":[127],"that":[128],"SoRFT":[129],"significantly":[130],"enhances":[131],"performance,":[133],"improves":[134],"generalization,":[136],"provides":[138],"cost-efficient":[140],"alternative":[141],"models.":[144]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":4}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}