{"id":"https://openalex.org/W7163154446","doi":"https://doi.org/10.48550/arxiv.2606.01400","title":"Consistent and Distinctive: LLM Benchmark Efficiency via Maximum Independent Set Prompt Selection on Similarity Graphs","display_name":"Consistent and Distinctive: LLM Benchmark Efficiency via Maximum Independent Set Prompt Selection on Similarity Graphs","publication_year":2026,"publication_date":"2026-05-31","ids":{"openalex":"https://openalex.org/W7163154446","doi":"https://doi.org/10.48550/arxiv.2606.01400"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.01400","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.01400","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.01400","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5128351043","display_name":"Denica Kjorvezir","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kjorvezir, Denica","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111191723","display_name":"Marko Djukanovi\u0107","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Djukanovi\u0107, Marko","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034185097","display_name":"Ana Gjorgjevikj","orcid":"https://orcid.org/0000-0002-5135-7718"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gjorgjevikj, Ana","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066396028","display_name":"Gjorgjina Cenikj","orcid":"https://orcid.org/0000-0002-2723-0821"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cenikj, Gjorgjina","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5082115266","display_name":"Tome Eftimov","orcid":"https://orcid.org/0000-0001-7330-1902"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Eftimov, Tome","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":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":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.2989000082015991,"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/T10028","display_name":"Topic Modeling","score":0.2989000082015991,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.22759999334812164,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.11999999731779099,"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":0.722100019454956},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.6255000233650208},{"id":"https://openalex.org/keywords/percentile","display_name":"Percentile","score":0.6061999797821045},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5430999994277954},{"id":"https://openalex.org/keywords/similarity","display_name":"Similarity (geometry)","score":0.5159000158309937},{"id":"https://openalex.org/keywords/ranking","display_name":"Ranking (information retrieval)","score":0.5085999965667725},{"id":"https://openalex.org/keywords/divergence","display_name":"Divergence (linguistics)","score":0.5081999897956848},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.48030000925064087}],"concepts":[{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.722100019454956},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.6255000233650208},{"id":"https://openalex.org/C122048520","wikidata":"https://www.wikidata.org/wiki/Q2913954","display_name":"Percentile","level":2,"score":0.6061999797821045},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5430999994277954},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.5159000158309937},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.5085999965667725},{"id":"https://openalex.org/C207390915","wikidata":"https://www.wikidata.org/wiki/Q1230525","display_name":"Divergence (linguistics)","level":2,"score":0.5081999897956848},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.48030000925064087},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.4691999852657318},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4643000066280365},{"id":"https://openalex.org/C122818955","wikidata":"https://www.wikidata.org/wiki/Q1060343","display_name":"Independent set","level":3,"score":0.444599986076355},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.4390999972820282},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.40860000252723694},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.37950000166893005},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.3774000108242035},{"id":"https://openalex.org/C148483581","wikidata":"https://www.wikidata.org/wiki/Q446488","display_name":"Feature selection","level":2,"score":0.32739999890327454},{"id":"https://openalex.org/C122123141","wikidata":"https://www.wikidata.org/wiki/Q176623","display_name":"Random variable","level":2,"score":0.3109999895095825},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.31049999594688416},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2985999882221222},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.29649999737739563},{"id":"https://openalex.org/C2639959","wikidata":"https://www.wikidata.org/wiki/Q1344778","display_name":"Distance measures","level":2,"score":0.2953000068664551},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.29510000348091125},{"id":"https://openalex.org/C22029948","wikidata":"https://www.wikidata.org/wiki/Q45089","display_name":"Dice","level":2,"score":0.2743000090122223},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.27390000224113464},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2621999979019165},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.25060001015663147}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.01400","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.01400","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.01400","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.01400","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":"Preprint"},"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":{"Evaluating":[0],"large":[1],"language":[2],"models":[3,20],"(LLMs)":[4],"across":[5,65],"comprehensive":[6],"benchmarks":[7,77,160],"is":[8,109],"expensive":[9],"and":[10,42,75,159],"time-consuming.":[11],"We":[12,56],"propose":[13],"a":[14,24,38,51],"graph-based":[15],"prompt":[16,136],"selection":[17,91],"framework":[18],"that":[19,89,100],"each":[21],"benchmark":[22,145],"as":[23,167],"similarity":[25],"graph":[26],"--":[27,41,88,108],"nodes":[28],"are":[29],"prompts":[30],"connected":[31],"if":[32],"their":[33],"embedding-space":[34],"distance":[35,70],"falls":[36],"above":[37],"configurable":[39],"threshold":[40],"applies":[43],"Maximum":[44],"Independent":[45],"Set":[46],"(MIS)":[47],"algorithms":[48],"to":[49],"select":[50],"maximally":[52],"diverse,":[53],"non-redundant":[54],"subset.":[55],"evaluate":[57],"four":[58,76],"MIS":[59],"solvers":[60],"(CPLEX,":[61],"GREEDY,":[62],"Online-MIS,":[63],"ReduMIS)":[64],"six":[66,72],"embedding":[67],"models,":[68],"three":[69],"measures,":[71],"percentile":[73,130],"thresholds,":[74],"(GPQA,":[78,161],"IFEval,":[79],"MMLU-Pro,":[80],"Omni-MATH)":[81],"covering":[82],"66":[83],"LLMs.":[84],"Our":[85],"central":[86],"hypothesis":[87],"repeated":[90],"under":[92],"different":[93],"random":[94],"seeds":[95],"yields":[96],"consistent":[97],"LLM":[98],"rankings":[99],"may":[101],"also":[102],"differ":[103],"from":[104,142],"the":[105,143,168],"full-benchmark":[106],"baseline":[107],"strongly":[110],"confirmed:":[111],"Kendall's":[112],"$W":[113,122],"\\geq":[114],"0.90$":[115],"in":[116,149],"99.2\\%":[117],"of":[118,152],"stochastic":[119],"configurations":[120],"(mean":[121],"=":[123],"0.997":[124],"\\pm":[125],"0.008$),":[126],"while":[127],"at":[128,155],"higher":[129],"thresholds":[131,157],"selected":[132],"subsets":[133],"achieve":[134],"25--48\\%":[135],"reduction":[137],"on":[138],"average.":[139],"Ranking":[140],"divergence":[141],"full":[144],"($\u03c1<":[146],"0.95$)":[147],"occurs":[148],"only":[150],"15.95\\%":[151],"configurations,":[153],"concentrated":[154],"low":[156],"($p_{10}$--$p_{20}$)":[158],"IFEval),":[162],"identifying":[163],"overly":[164],"dense":[165],"graphs":[166],"primary":[169],"failure":[170],"mode.":[171]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-03T00:00:00"}