{"id":"https://openalex.org/W4416035949","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.68","title":"Large Language Model Evaluation via Matrix Nuclear-Norm","display_name":"Large Language Model Evaluation via Matrix Nuclear-Norm","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416035949","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.68"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.68","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.68","pdf_url":"https://aclanthology.org/2025.findings-emnlp.68.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":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.68.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5113409309","display_name":"Y. S. Li","orcid":"https://orcid.org/0009-0003-6542-2239"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yahan Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083820813","display_name":"Tingyu Xia","orcid":"https://orcid.org/0000-0002-4785-9969"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tingyu Xia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102024536","display_name":"Yuan Wu","orcid":"https://orcid.org/0000-0001-6289-5872"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuan Wu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5049349192","display_name":"Yi Chang","orcid":"https://orcid.org/0000-0001-8542-5937"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yi Chang","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":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1306","last_page":"1323"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.12639999389648438,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.12639999389648438,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.05429999902844429,"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/T14347","display_name":"Big Data and Digital Economy","score":0.051600001752376556,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/matrix","display_name":"Matrix (chemical analysis)","score":0.38339999318122864},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.3012999892234802},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.24789999425411224},{"id":"https://openalex.org/keywords/algebra-over-a-field","display_name":"Algebra over a field","score":0.24469999969005585}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5777000188827515},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.42089998722076416},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4011000096797943},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.38339999318122864},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3012999892234802},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.28540000319480896},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.28299999237060547},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.28130000829696655},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.24789999425411224},{"id":"https://openalex.org/C136119220","wikidata":"https://www.wikidata.org/wiki/Q1000660","display_name":"Algebra over a field","level":2,"score":0.24469999969005585}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.68","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.68","pdf_url":"https://aclanthology.org/2025.findings-emnlp.68.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.68","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.68","pdf_url":"https://aclanthology.org/2025.findings-emnlp.68.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":[{"id":"https://openalex.org/G6592930967","display_name":null,"funder_award_id":"2023YFF0905400","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"},{"id":"https://openalex.org/G8336371671","display_name":null,"funder_award_id":"U2341229","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320322174","display_name":"People's Government of Jilin Province","ror":"https://ror.org/02fzqav45"},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null},{"id":"https://openalex.org/F4320335787","display_name":"Fundamental Research Funds for the Central Universities","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416035949.pdf","grobid_xml":"https://content.openalex.org/works/W4416035949.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"As":[0],"large":[1],"language":[2],"models":[3,37,164],"(LLMs)":[4],"continue":[5],"to":[6,17,39,66,83,95,115,132,148],"evolve,":[7],"efficient":[8,184],"evaluation":[9],"metrics":[10,24],"are":[11,32],"vital":[12],"for":[13,35,123,139,186],"assessing":[14,187],"their":[15,40],"ability":[16],"compress":[18],"information":[19,107],"and":[20,88,119,170,183,195],"reduce":[21],"redundancy.While":[22],"traditional":[23],"like":[25,165],"Matrix":[26,57,127,137,177],"Entropy":[27,138],"offer":[28],"valuable":[29],"insights,":[30],"they":[31],"computationally":[33],"intensive":[34],"largescale":[36],"due":[38],"O(n":[41,116],"3":[42],")":[43,118],"time":[44,113],"complexity":[45,114],"with":[46,155,162],"Singular":[47],"Value":[48],"Decomposition":[49],"(SVD).To":[50],"mitigate":[51],"this":[52],"issue,":[53],"we":[54,101],"introduce":[55],"the":[56,68,91,98,105,112,121,126,140],"Nuclear-Norm,":[58],"which":[59],"not":[60],"only":[61],"serves":[62],"as":[63,143,158],"a":[64,77,180,191],"metric":[65],"quantify":[67],"data":[69],"compression":[70,108],"proficiency":[71],"of":[72,80],"LLM":[73],"but":[74],"also":[75],"provides":[76],"convex":[78],"approximation":[79],"matrix":[81],"rank":[82],"capture":[84],"both":[85],"predictive":[86],"discriminability":[87],"diversity.By":[89],"employing":[90],"L":[92],"1,2":[93],"-norm":[94],"further":[96],"approximate":[97],"nuclear":[99],"norm,":[100],"can":[102],"effectively":[103],"assess":[104],"model's":[106],"capabilities.This":[109],"approach":[110],"reduces":[111],"2":[117],"eliminates":[120],"need":[122],"SVD":[124],"computation.Consequently,":[125],"Nuclear-Norm":[128,178],"achieves":[129],"speeds":[130],"8":[131],"24":[133],"times":[134],"faster":[135],"than":[136],"Cerebras-GPT":[141],"model":[142,171],"sizes":[144],"increase":[145],"from":[146],"111M":[147],"6.7B.This":[149],"performance":[150],"gap":[151],"becomes":[152],"more":[153],"pronounced":[154],"larger":[156],"models,":[157],"validated":[159],"in":[160],"tests":[161],"other":[163],"Pythia.Additionally,":[166],"evaluations":[167],"on":[168],"benchmarks":[169],"responses":[172],"confirm":[173],"that":[174],"our":[175],"proposed":[176],"is":[179],"reliable,":[181],"scalable,":[182],"tool":[185],"LLMs'":[188],"performance,":[189],"striking":[190],"balance":[192],"between":[193],"accuracy":[194],"computational":[196],"efficiency.":[197]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-11-08T00:00:00"}