{"id":"https://openalex.org/W7090779509","doi":"https://doi.org/10.48550/arxiv.2510.10457","title":"Rethinking LLM Evaluation: Can We Evaluate LLMs with 200x Less Data?","display_name":"Rethinking LLM Evaluation: Can We Evaluate LLMs with 200x Less Data?","publication_year":2025,"publication_date":"2025-10-12","ids":{"openalex":"https://openalex.org/W7090779509","doi":"https://doi.org/10.48550/arxiv.2510.10457"},"language":"en","primary_location":{"id":"doi:10.48550/arxiv.2510.10457","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2510.10457","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2510.10457","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Wang, Shaobo","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Wang, Shaobo","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Wang, Cong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Cong","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Fu, Wenjie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fu, Wenjie","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Min, Yue","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Min, Yue","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Feng, Mingquan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Feng, Mingquan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Guan, Isabel","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guan, Isabel","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Hu, Xuming","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Xuming","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"He, Conghui","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Conghui","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Wang, Cunxiang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Cunxiang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Yang, Kexin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Kexin","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Ren, Xingzhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ren, Xingzhang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Huang, Fei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Fei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Liu, Dayiheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Dayiheng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Zhang, Linfeng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Linfeng","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":14,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":true,"primary_topic":{"id":"https://openalex.org/T12535","display_name":"Machine Learning and Data Classification","score":0.17679999768733978,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.17679999768733978,"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/T11652","display_name":"Imbalanced Data Classification Techniques","score":0.09300000220537186,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.0771000012755394,"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/benchmark","display_name":"Benchmark (surveying)","score":0.8155999779701233},{"id":"https://openalex.org/keywords/redundancy","display_name":"Redundancy (engineering)","score":0.6776000261306763},{"id":"https://openalex.org/keywords/ranking","display_name":"Ranking (information retrieval)","score":0.6394000053405762},{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.5335999727249146},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.3296000063419342},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.313400000333786}],"concepts":[{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.8155999779701233},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.6776000261306763},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.6394000053405762},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.5335999727249146},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5224000215530396},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5113999843597412},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4018000066280365},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3889000117778778},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.3296000063419342},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.313400000333786},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.2793999910354614},{"id":"https://openalex.org/C181243257","wikidata":"https://www.wikidata.org/wiki/Q1693522","display_name":"Sample mean and sample covariance","level":3,"score":0.26910001039505005},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.26190000772476196},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.25099998712539673}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2510.10457","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2510.10457","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2510.10457","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2510.10457","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"As":[0],"the":[1,82,105,136,144],"demand":[2],"for":[3],"comprehensive":[4],"evaluations":[5],"of":[6,59,84,107,146],"diverse":[7],"model":[8],"capabilities":[9],"steadily":[10],"increases,":[11],"benchmark":[12,60,75,138],"suites":[13],"have":[14],"correspondingly":[15],"grown":[16],"significantly":[17],"in":[18,23],"scale.":[19],"Despite":[20],"notable":[21],"advances":[22],"redundancy":[24,61],"reduction":[25],"and":[26,43,62,111,129,156],"subset-level":[27],"performance":[28],"prediction,":[29],"a":[30,56,94],"systematic":[31],"framework":[32,96],"that":[33,68],"effectively":[34],"integrates":[35],"these":[36],"methods":[37],"to":[38,116],"ensure":[39],"both":[40],"prediction":[41],"accuracy":[42],"ranking":[44,145,159],"consistency":[45],"is":[46],"still":[47],"largely":[48],"elusive.":[49],"In":[50,133],"this":[51],"paper,":[52],"we":[53,73,90],"first":[54],"perform":[55],"sample-level":[57],"analysis":[58],"identify":[63],"several":[64],"highly":[65],"similar":[66],"samples":[67],"can":[69],"be":[70],"eliminated.":[71],"Besides,":[72],"frame":[74],"compression":[76,123],"as":[77],"an":[78,98],"optimization":[79],"problem":[80],"with":[81,125],"aim":[83],"score":[85],"reconstruction.":[86],"Building":[87],"on":[88,135],"these,":[89],"then":[91],"propose":[92],"EssenceBench,":[93],"coarse-to-fine":[95],"utilizing":[97],"iterative":[99],"Genetic":[100],"Algorithm":[101],"(GA),":[102],"which":[103],"takes":[104],"advantages":[106],"fitness-based":[108],"subset":[109],"search":[110],"attribution-based":[112],"sample":[113],"search.":[114],"Compared":[115],"previous":[117],"methods,":[118],"our":[119,141],"approach":[120],"yields":[121],"superior":[122],"results":[124],"lower":[126],"reconstruction":[127],"error":[128],"markedly":[130],"higher":[131],"efficiency.":[132],"particular,":[134],"HellaSwag":[137],"(10K":[139],"samples),":[140],"method":[142],"preserves":[143],"all":[147],"models":[148],"shifting":[149,161],"within":[150,162],"5%":[151,163],"using":[152,164],"25x":[153],"fewer":[154,167],"samples,":[155],"achieves":[157],"95%":[158],"preservation":[160],"only":[165],"200x":[166],"samples.":[168]},"counts_by_year":[],"updated_date":"2025-11-06T06:51:31.235846","created_date":"2025-10-15T00:00:00"}