{"id":"https://openalex.org/W4412886766","doi":"https://doi.org/10.18653/v1/2025.acl-long.147","title":"TRACT: Regression-Aware Fine-tuning Meets Chain-of-Thought Reasoning for LLM-as-a-Judge","display_name":"TRACT: Regression-Aware Fine-tuning Meets Chain-of-Thought Reasoning for LLM-as-a-Judge","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412886766","doi":"https://doi.org/10.18653/v1/2025.acl-long.147"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.147","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.147","pdf_url":"https://aclanthology.org/2025.acl-long.147.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.147.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5071482462","display_name":"Cheng-Han Chiang","orcid":null},"institutions":[{"id":"https://openalex.org/I112289208","display_name":"United Nations Children's Fund","ror":"https://ror.org/02dg0pv02","country_code":"US","type":"funder","lineage":["https://openalex.org/I112289208","https://openalex.org/I1286959531"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Cheng-Han Chiang","raw_affiliation_strings":["NTU GICE Taiwan NTU GICE Taiwan Google Research New York"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NTU GICE Taiwan NTU GICE Taiwan Google Research New York","institution_ids":["https://openalex.org/I112289208"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040508737","display_name":"Hung-yi Lee","orcid":"https://orcid.org/0000-0002-9654-5747"},"institutions":[{"id":"https://openalex.org/I112289208","display_name":"United Nations Children's Fund","ror":"https://ror.org/02dg0pv02","country_code":"US","type":"funder","lineage":["https://openalex.org/I112289208","https://openalex.org/I1286959531"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hung-yi Lee","raw_affiliation_strings":["NTU GICE Taiwan NTU GICE Taiwan Google Research New York"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NTU GICE Taiwan NTU GICE Taiwan Google Research New York","institution_ids":["https://openalex.org/I112289208"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5083277438","display_name":"Micha\u0142 \u0141ukasik","orcid":null},"institutions":[{"id":"https://openalex.org/I112289208","display_name":"United Nations Children's Fund","ror":"https://ror.org/02dg0pv02","country_code":"US","type":"funder","lineage":["https://openalex.org/I112289208","https://openalex.org/I1286959531"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Michal Lukasik","raw_affiliation_strings":["NTU GICE Taiwan NTU GICE Taiwan Google Research New York"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NTU GICE Taiwan NTU GICE Taiwan Google Research New York","institution_ids":["https://openalex.org/I112289208"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I112289208"],"apc_list":null,"apc_paid":null,"fwci":24.9416,"has_fulltext":true,"cited_by_count":4,"citation_normalized_percentile":{"value":0.99171687,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"2934","last_page":"2952"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13643","display_name":"Artificial Intelligence in Law","score":0.9065999984741211,"subfield":{"id":"https://openalex.org/subfields/3320","display_name":"Political Science and International Relations"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T13643","display_name":"Artificial Intelligence in Law","score":0.9065999984741211,"subfield":{"id":"https://openalex.org/subfields/3320","display_name":"Political Science and International Relations"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.9035000205039978,"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":0.6397070288658142},{"id":"https://openalex.org/keywords/chain","display_name":"Chain (unit)","score":0.46042725443840027},{"id":"https://openalex.org/keywords/regression","display_name":"Regression","score":0.44790810346603394},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3841100335121155},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.255110502243042},{"id":"https://openalex.org/keywords/psychoanalysis","display_name":"Psychoanalysis","score":0.07446259260177612}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6397070288658142},{"id":"https://openalex.org/C199185054","wikidata":"https://www.wikidata.org/wiki/Q552299","display_name":"Chain (unit)","level":2,"score":0.46042725443840027},{"id":"https://openalex.org/C83546350","wikidata":"https://www.wikidata.org/wiki/Q1139051","display_name":"Regression","level":2,"score":0.44790810346603394},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3841100335121155},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.255110502243042},{"id":"https://openalex.org/C11171543","wikidata":"https://www.wikidata.org/wiki/Q41630","display_name":"Psychoanalysis","level":1,"score":0.07446259260177612},{"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.acl-long.147","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.147","pdf_url":"https://aclanthology.org/2025.acl-long.147.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.147","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.147","pdf_url":"https://aclanthology.org/2025.acl-long.147.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":[],"funders":[{"id":"https://openalex.org/F4320318804","display_name":"Delta Electronics","ror":"https://ror.org/04s3g5933"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412886766.pdf","grobid_xml":"https://content.openalex.org/works/W4412886766.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":{"The":[0],"LLM-as-a-judge":[1,26,134],"paradigm":[2],"uses":[3],"large":[4],"language":[5],"models":[6],"(LLMs)":[7],"for":[8,25,31,60,89,98],"automated":[9],"text":[10,18],"evaluation,":[11],"which":[12,33],"assigns":[13],"a":[14,73,107],"score":[15,39,61,100],"to":[16,112],"the":[17,35,86,91,95,99,118,122,126,149],"based":[19],"on":[20],"some":[21],"scoring":[22],"rubrics.Existing":[23],"methods":[24],"use":[27],"cross-entropy":[28],"(CE)":[29],"loss":[30,88,97],"fine-tuning,":[32],"neglects":[34],"numeric":[36],"nature":[37],"of":[38,46,83,103,151],"prediction.Recent":[40],"work":[41],"addresses":[42],"numerical":[43],"prediction":[44],"limitations":[45],"LLM":[47,109,120,127],"fine-tuning":[48,70],"through":[49],"regression-aware":[50,79,96],"finetuning,":[51],"which,":[52],"however,":[53],"does":[54],"not":[55],"consider":[56],"chain-of-thought":[57],"(CoT)":[58],"reasoning":[59,77,93],"prediction.In":[62],"this":[63],"paper,":[64],"we":[65,116],"introduce":[66],"TRACT":[67,84,141],"(Two-stage":[68],"Regression-Aware":[69],"with":[71,78],"CoT),":[72],"method":[74],"combining":[75],"CoT":[76,92],"training.The":[80],"training":[81],"objective":[82],"combines":[85],"CE":[87],"learning":[90],"and":[94,136],"prediction.TRACT":[101],"consists":[102],"two":[104,137],"stages:":[105],"first,":[106],"seed":[108,119],"is":[110],"fine-tuned":[111],"generate":[113],"CoTs;":[114],"next,":[115],"retrain":[117],"using":[121],"CoTs":[123],"generated":[124],"by":[125],"trained":[128],"in":[129,154],"stage":[130],"1.Experiments":[131],"across":[132],"four":[133],"datasets":[135],"LLMs":[138],"show":[139],"that":[140],"significantly":[142],"outperforms":[143],"existing":[144],"methods.Extensive":[145],"ablation":[146],"studies":[147],"validate":[148],"importance":[150],"each":[152],"component":[153],"TRACT.":[155],"1":[156]},"counts_by_year":[{"year":2026,"cited_by_count":4}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}