{"id":"https://openalex.org/W4416034522","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.1342","title":"Do Before You Judge: Self-Reference as a Pathway to Better LLM Evaluation","display_name":"Do Before You Judge: Self-Reference as a Pathway to Better LLM Evaluation","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034522","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.1342"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.1342","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1342","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1342.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.1342.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5022604221","display_name":"Wei-Hsiang Lin","orcid":"https://orcid.org/0000-0001-8177-5892"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei-Hsiang Lin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113392197","display_name":"Sheng-Lun Wei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sheng-Lun Wei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120308624","display_name":"Hen-Hsen Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hen-Hsen Huang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5000334344","display_name":"Hsin\u2010Hsi Chen","orcid":"https://orcid.org/0000-0001-9757-9423"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hsin-Hsi Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"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":0.4358743,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"24651","last_page":"24672"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11894","display_name":"Radiology practices and education","score":0.048900000751018524,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},"topics":[{"id":"https://openalex.org/T11894","display_name":"Radiology practices and education","score":0.048900000751018524,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T11314","display_name":"Clinical Laboratory Practices and Quality Control","score":0.019700000062584877,"subfield":{"id":"https://openalex.org/subfields/2737","display_name":"Physiology"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T11710","display_name":"Biomedical Text Mining and Ontologies","score":0.019700000062584877,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.2637999951839447},{"id":"https://openalex.org/keywords/medline","display_name":"MEDLINE","score":0.2386000007390976},{"id":"https://openalex.org/keywords/perspective","display_name":"Perspective (graphical)","score":0.2199999988079071},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.2092999964952469},{"id":"https://openalex.org/keywords/work","display_name":"Work (physics)","score":0.20829999446868896}],"concepts":[{"id":"https://openalex.org/C71924100","wikidata":"https://www.wikidata.org/wiki/Q11190","display_name":"Medicine","level":0,"score":0.49540001153945923},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.2980000078678131},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.2637999951839447},{"id":"https://openalex.org/C2779473830","wikidata":"https://www.wikidata.org/wiki/Q1540899","display_name":"MEDLINE","level":2,"score":0.2386000007390976},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.2199999988079071},{"id":"https://openalex.org/C177713679","wikidata":"https://www.wikidata.org/wiki/Q679690","display_name":"Intensive care medicine","level":1,"score":0.21240000426769257},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.2092999964952469},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.20829999446868896},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.20669999718666077},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.2004999965429306}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.1342","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1342","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1342.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.1342","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1342","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1342.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":[{"id":"https://openalex.org/F4320322724","display_name":"Ministry of Education, India","ror":"https://ror.org/048xjjh50"},{"id":"https://openalex.org/F4320331164","display_name":"National Science and Technology Council","ror":"https://ror.org/00wnb9798"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034522.pdf","grobid_xml":"https://content.openalex.org/works/W4416034522.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"LLM-as-Judge":[0],"frameworks":[1],"are":[2,49],"increasingly":[3],"popular":[4],"for":[5,102],"AI":[6],"evaluation,":[7],"yet":[8],"research":[9],"findings":[10],"on":[11,40],"the":[12,41,59,82],"relationship":[13,24],"between":[14,84],"models'":[15],"generation":[16,85],"and":[17,33,86,97],"judgment":[18,87],"abilities":[19],"remain":[20],"inconsistent.We":[21],"investigate":[22],"this":[23],"through":[25],"systematic":[26],"dataset-and":[27],"instance-level":[28],"analyses":[29,46],"across":[30],"11":[31],"models":[32],"21":[34],"diverse":[35],"tasks.Despite":[36],"both":[37],"capabilities":[38],"relying":[39],"same":[42],"underlying":[43],"knowledge,":[44],"our":[45],"reveal":[47],"they":[48],"only":[50],"weakly":[51],"correlated,":[52],"primarily":[53],"due":[54],"to":[55,58,93],"LLMs'":[56],"sensitivity":[57],"responses":[60],"being":[61],"judged.To":[62],"address":[63],"this,":[64],"we":[65],"propose":[66],"a":[67,73,90,99],"self-reference-guided":[68],"evaluation":[69,106],"strategy":[70],"that":[71],"leverages":[72],"model's":[74],"own":[75],"answers":[76],"as":[77],"references.This":[78],"approach":[79],"significantly":[80],"strengthens":[81],"correlation":[83],"abilities,":[88],"offering":[89],"practical":[91],"path":[92],"align":[94],"these":[95],"skills":[96],"providing":[98],"reliable":[100],"proxy":[101],"model":[103],"selection":[104],"in":[105],"tasks.":[107]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-11-08T00:00:00"}