{"id":"https://openalex.org/W4416034132","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.1259","title":"Improving LLM-as-a-Judge Inference with the Judgment Distribution","display_name":"Improving LLM-as-a-Judge Inference with the Judgment Distribution","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034132","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.1259"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.1259","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1259","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1259.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: EMNLP 2025","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.1259.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5020284776","display_name":"Victor Wang","orcid":"https://orcid.org/0000-0001-9633-1506"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Victor Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014048260","display_name":"M Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Michael JQ Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5035142405","display_name":"Eunsol Choi","orcid":"https://orcid.org/0000-0003-3607-9104"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Eunsol Choi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.35047163,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"23173","last_page":"23199"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13643","display_name":"Artificial Intelligence in Law","score":0.4575999975204468,"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":0.4575999975204468,"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/T11762","display_name":"Law, Economics, and Judicial Systems","score":0.09309999644756317,"subfield":{"id":"https://openalex.org/subfields/2002","display_name":"Economics and Econometrics"},"field":{"id":"https://openalex.org/fields/20","display_name":"Economics, Econometrics and Finance"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T13583","display_name":"Jury Decision Making Processes","score":0.053599998354911804,"subfield":{"id":"https://openalex.org/subfields/3308","display_name":"Law"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/distribution","display_name":"Distribution (mathematics)","score":0.4165000021457672},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.38850000500679016},{"id":"https://openalex.org/keywords/fiducial-inference","display_name":"Fiducial inference","score":0.295199990272522},{"id":"https://openalex.org/keywords/probability-distribution","display_name":"Probability distribution","score":0.2775999903678894},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.2709999978542328}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5145000219345093},{"id":"https://openalex.org/C110121322","wikidata":"https://www.wikidata.org/wiki/Q865811","display_name":"Distribution (mathematics)","level":2,"score":0.4165000021457672},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4032000005245209},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.38850000500679016},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3578000068664551},{"id":"https://openalex.org/C95167961","wikidata":"https://www.wikidata.org/wiki/Q4483495","display_name":"Fiducial inference","level":5,"score":0.295199990272522},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.29350000619888306},{"id":"https://openalex.org/C149441793","wikidata":"https://www.wikidata.org/wiki/Q200726","display_name":"Probability distribution","level":2,"score":0.2775999903678894},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.27250000834465027},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2709999978542328},{"id":"https://openalex.org/C134261354","wikidata":"https://www.wikidata.org/wiki/Q938438","display_name":"Statistical inference","level":2,"score":0.26930001378059387},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.2621999979019165},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.25130000710487366}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.1259","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1259","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1259.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: EMNLP 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.1259","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1259","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1259.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: EMNLP 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/W4416034132.pdf","grobid_xml":"https://content.openalex.org/works/W4416034132.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Using":[0],"language":[1],"models":[2],"to":[3,18,129],"scalably":[4],"approximate":[5],"human":[6],"preferences":[7,83],"on":[8],"text":[9,132],"quality":[10],"(LLMas-a-judge)":[11],"has":[12],"become":[13],"a":[14,44],"standard":[15],"practice":[16],"applicable":[17],"many":[19],"tasks.A":[20],"judgment":[21,41,60,85,114],"is":[22],"often":[23,94,116],"extracted":[24],"from":[25,84],"the":[26,56,59,65,110,113,131],"judge's":[27],"textual":[28],"output":[29,124],"alone,":[30],"typically":[31],"with":[32,101],"greedy":[33],"decoding.However,":[34],"LLM":[35],"judges":[36],"naturally":[37],"provide":[38],"distributions":[39],"over":[40],"tokens,":[42],"inviting":[43],"breadth":[45],"of":[46,58,81,112],"inference":[47],"methods":[48,80,90],"for":[49],"extracting":[50],"fine-grained":[51],"preferences.We":[52],"find":[53,88],"that":[54,89,106,121],"taking":[55,64],"mean":[57],"distribution":[61],"consistently":[62],"outperforms":[63],"mode":[66],"(i.e.greedy":[67],"decoding)":[68],"in":[69],"all":[70],"evaluation":[71],"settings":[72],"(i.e.pointwise,":[73],"pairwise,":[74],"and":[75,87],"listwise).We":[76],"further":[77],"explore":[78],"novel":[79],"deriving":[82],"distributions,":[86],"incorporating":[91],"risk":[92],"aversion":[93],"improve":[95],"performance.Lastly,":[96],"we":[97],"analyze":[98],"LLM-as-a-judge":[99],"paired":[100],"chain-of-thought":[102],"(CoT)":[103],"prompting,":[104],"showing":[105],"CoT":[107],"can":[108],"collapse":[109],"spread":[111],"distribution,":[115],"harming":[117],"performance.Our":[118],"findings":[119],"show":[120],"leveraging":[122],"distributional":[123],"improves":[125],"LLM-as-a-judge,":[126],"as":[127],"opposed":[128],"using":[130],"interface":[133],"alone.":[134]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-11-08T00:00:00"}