{"id":"https://openalex.org/W7220737195","doi":"https://doi.org/10.48550/arxiv.2610.05058","title":"A Statistical Inference Framework for PMI Estimation and SGNS Word Embeddings","display_name":"A Statistical Inference Framework for PMI Estimation and SGNS Word Embeddings","publication_year":2026,"publication_date":"2026-10-04","ids":{"openalex":"https://openalex.org/W7220737195","doi":"https://doi.org/10.48550/arxiv.2610.05058"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.05058","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.05058","pdf_url":"https://arxiv.org/pdf/2610.05058","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2610.05058","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5152832091","display_name":"Zhongqi Fan","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I12615008","display_name":"Beijing Normal-Hong Kong Baptist University","ror":"https://ror.org/04snvc712","country_code":"CN","type":"education","lineage":["https://openalex.org/I12615008"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Fan, Zhongqi","raw_affiliation_strings":["Beijing Normal-Hong Kong Baptist University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Beijing Normal-Hong Kong Baptist University","institution_ids":["https://openalex.org/I12615008"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5152832091"],"corresponding_institution_ids":["https://openalex.org/I12615008"],"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":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":0.9794999957084656,"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":0.9794999957084656,"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":0.008899999782443047,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.0017000000225380063,"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/pointwise-mutual-information","display_name":"pointwise mutual information","score":0.9860000014305115},{"id":"https://openalex.org/keywords/word-embeddings","display_name":"word embeddings","score":0.9850000143051147},{"id":"https://openalex.org/keywords/estimator-consistency","display_name":"estimator consistency","score":0.5740000009536743},{"id":"https://openalex.org/keywords/asymptotic-unbiasedness","display_name":"asymptotic unbiasedness","score":0.9660000205039978},{"id":"https://openalex.org/keywords/asymptotic-normality","display_name":"asymptotic normality","score":0.9990000128746033},{"id":"https://openalex.org/keywords/delta-method","display_name":"delta method","score":0.8640000224113464},{"id":"https://openalex.org/keywords/variance-decomposition","display_name":"variance decomposition","score":0.9599999785423279},{"id":"https://openalex.org/keywords/stochastic-approximation","display_name":"stochastic approximation","score":0.671999990940094},{"id":"https://openalex.org/keywords/word-association","display_name":"word association","score":0.6650000214576721},{"id":"https://openalex.org/keywords/brown-corpus","display_name":"Brown Corpus","score":0.7699999809265137},{"id":"https://openalex.org/keywords/word-analogy","display_name":"word analogy","score":0.8799999952316284}],"concepts":[{"id":"https://openalex.org/C2777984123","wikidata":"https://www.wikidata.org/wiki/Q9248237","display_name":"Pointwise","level":2,"score":0.6830999851226807},{"id":"https://openalex.org/C971699","wikidata":"https://www.wikidata.org/wiki/Q1132714","display_name":"Delta method","level":3,"score":0.5824000239372253},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.5656999945640564},{"id":"https://openalex.org/C134261354","wikidata":"https://www.wikidata.org/wiki/Q938438","display_name":"Statistical inference","level":2,"score":0.5404000282287598},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.5083000063896179},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5078999996185303},{"id":"https://openalex.org/C7797323","wikidata":"https://www.wikidata.org/wiki/Q3798612","display_name":"Pointwise mutual information","level":3,"score":0.48500001430511475},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.4844000041484833},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.46779999136924744},{"id":"https://openalex.org/C65778772","wikidata":"https://www.wikidata.org/wiki/Q12345341","display_name":"Asymptotic distribution","level":3,"score":0.4465000033378601},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4083000123500824},{"id":"https://openalex.org/C119047807","wikidata":"https://www.wikidata.org/wiki/Q752718","display_name":"Asymptotic analysis","level":2,"score":0.4041000008583069},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.39959999918937683},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.39250001311302185},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.38519999384880066},{"id":"https://openalex.org/C207609745","wikidata":"https://www.wikidata.org/wiki/Q4944086","display_name":"Bootstrapping (finance)","level":2,"score":0.3662000000476837},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.3483000099658966},{"id":"https://openalex.org/C87007009","wikidata":"https://www.wikidata.org/wiki/Q210832","display_name":"Statistical hypothesis testing","level":2,"score":0.3379000127315521},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3370000123977661},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33500000834465027},{"id":"https://openalex.org/C13355873","wikidata":"https://www.wikidata.org/wiki/Q2920850","display_name":"Connection (principal bundle)","level":2,"score":0.3305000066757202},{"id":"https://openalex.org/C124681953","wikidata":"https://www.wikidata.org/wiki/Q339062","display_name":"Decomposition","level":2,"score":0.32670000195503235},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.2971999943256378},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.2879999876022339},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.28790000081062317},{"id":"https://openalex.org/C2777601683","wikidata":"https://www.wikidata.org/wiki/Q6499736","display_name":"Vocabulary","level":2,"score":0.28439998626708984},{"id":"https://openalex.org/C164660894","wikidata":"https://www.wikidata.org/wiki/Q2037833","display_name":"Piecewise","level":2,"score":0.2791000008583069},{"id":"https://openalex.org/C167928553","wikidata":"https://www.wikidata.org/wiki/Q1376021","display_name":"Estimation theory","level":2,"score":0.2687999904155731},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.26809999346733093},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.26589998602867126},{"id":"https://openalex.org/C96250715","wikidata":"https://www.wikidata.org/wiki/Q965330","display_name":"Estimation","level":2,"score":0.26440000534057617},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.2624000012874603},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2603999972343445},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.25540000200271606},{"id":"https://openalex.org/C52622258","wikidata":"https://www.wikidata.org/wiki/Q131222","display_name":"Information theory","level":2,"score":0.2540999948978424},{"id":"https://openalex.org/C8272713","wikidata":"https://www.wikidata.org/wiki/Q176737","display_name":"Stochastic process","level":2,"score":0.2538999915122986},{"id":"https://openalex.org/C22789450","wikidata":"https://www.wikidata.org/wiki/Q420904","display_name":"Singular value decomposition","level":2,"score":0.25200000405311584}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.05058","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.05058","pdf_url":"https://arxiv.org/pdf/2610.05058","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2610.05058","doi":"https://doi.org/10.48550/arxiv.2610.05058","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.05058","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.05058","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.05058","pdf_url":"https://arxiv.org/pdf/2610.05058","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7220737195.pdf","grobid_xml":"https://content.openalex.org/works/W7220737195.grobid-xml"},"referenced_works_count":6,"referenced_works":["https://openalex.org/W1994616650","https://openalex.org/W2084053452","https://openalex.org/W2086161653","https://openalex.org/W3044617930","https://openalex.org/W4387724891","https://openalex.org/W4399379323"],"related_works":["https://openalex.org/W4396744270","https://openalex.org/W2125031621","https://openalex.org/W4282982859","https://openalex.org/W2294865516","https://openalex.org/W2968960671","https://openalex.org/W2252218921","https://openalex.org/W4416776198","https://openalex.org/W2161732531","https://openalex.org/W3110014267","https://openalex.org/W2903988268"],"abstract_inverted_index":{"Pointwise":[0],"Mutual":[1],"Information":[2],"(PMI)":[3],"is":[4,18],"a":[5,20,25,30,50,87,144],"core":[6],"measure":[7],"of":[8,35,69],"testing":[9],"word":[10],"association,":[11],"and":[12,32,43,56,66,143,184,191,198],"Skip-gram":[13],"with":[14],"Negative":[15],"Sampling":[16],"(SGNS)":[17],"essentially":[19],"method":[21,106],"that":[22,94,118],"implicitly":[23],"factorizes":[24],"shifted":[26],"PMI":[27,39,54,72,92,126,136],"matrix.":[28],"However,":[29],"systematic":[31,148],"well-rounded":[33],"characterization":[34],"finite-sample":[36],"uncertainty":[37],"in":[38],"estimation":[40,55,93],"remains":[41],"absent":[42],"imperative":[44],"to":[45,59,160],"venture":[46],"into.":[47],"We":[48,61],"provide":[49],"statistical":[51],"framework":[52],"for":[53,90,195],"its":[57,75,171],"connection":[58],"SGNS.":[60],"prove":[62],"consistency,":[63],"asymptotic":[64,67,189],"unbiasedness,":[65],"normality":[68],"the":[70,78,104,111,123,154,162,177,181,186],"empirical":[71,131],"estimator,":[73],"derive":[74],"variance":[76,88,97,163],"via":[77],"Delta":[79,105],"method,":[80],"and,":[81],"applying":[82],"stochastic":[83],"approximation":[84],"theory,":[85],"obtain":[86],"decomposition":[89,164],"SGNS-based":[91],"separates":[95],"data":[96],"from":[98,122],"optimization":[99],"variance.":[100],"Simulation":[101],"experiments":[102,109],"validate":[103,161],"approximation.":[107],"Real-data":[108],"on":[110],"Brown":[112],"Corpus":[113],"(d":[114],"=":[115],"100)":[116],"reveal":[117],"SGNS":[119],"systematically":[120],"deviates":[121],"theoretical":[124,172],"relationship":[125,132],"+":[127],"log":[128,140],"K.":[129],"The":[130,158],"shows":[133],"an":[134,138],"attenuated":[135],"coefficient,":[137],"amplified":[139],"K":[141],"effect,":[142],"positive":[145],"intercept,":[146],"indicating":[147],"bias.":[149],"Word":[150],"analogy":[151],"validation":[152],"confirms":[153],"models":[155],"are":[156],"effective.":[157],"failure":[159],"under":[165],"low-dimensional":[166],"conditions":[167],"does":[168],"not":[169],"diminish":[170],"value;":[173],"rather,":[174],"it":[175],"identifies":[176],"unbiasedness":[178],"assumption":[179],"as":[180],"key":[182],"bottleneck":[183],"clarifies":[185],"gap":[187],"between":[188],"theory":[190],"practice,":[192],"providing":[193],"implications":[194],"both":[196],"practice":[197],"theory.":[199]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-07T00:00:00"}