{"id":"https://openalex.org/W7160950339","doi":"https://doi.org/10.48550/arxiv.2605.08764","title":"Anchoring the Eigengap: Cross-Modal Spectral Stabilization for Sample-Efficient Representation Learning","display_name":"Anchoring the Eigengap: Cross-Modal Spectral Stabilization for Sample-Efficient Representation Learning","publication_year":2026,"publication_date":"2026-05-09","ids":{"openalex":"https://openalex.org/W7160950339","doi":"https://doi.org/10.48550/arxiv.2605.08764"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.08764","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08764","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":null,"license_id":null,"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.2605.08764","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5089189717","display_name":"Nikhil J. Dhinagar","orcid":"https://orcid.org/0000-0003-2424-4854"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dhinagar, Nikhil J.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002367969","display_name":"Vidhi Chhatbar","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chhatbar, Vidhi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5009543931","display_name":"Chirag Jagad","orcid":"https://orcid.org/0000-0002-6119-3956"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jagad, Chirag","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101772582","display_name":"Pavithra Senthilkumar","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Senthilkumar, Pavithra","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135923578","display_name":"Sophia I. Thomopoulos","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Thomopoulos, Sophia I.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034197000","display_name":"Mahir H. Khan","orcid":"https://orcid.org/0000-0002-7215-9585"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Khan, Mahir H.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133592833","display_name":"Sook-Lei Liew","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liew, Sook-Lei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135963069","display_name":"the ENIGMA-Stroke Recovery Working Group","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Group, the ENIGMA-Stroke Recovery Working","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135996634","display_name":"Paul M. Thompson","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Thompson, Paul M.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":9,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.28870001435279846,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.28870001435279846,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.15809999406337738,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.09570000320672989,"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/overfitting","display_name":"Overfitting","score":0.5091999769210815},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.4916999936103821},{"id":"https://openalex.org/keywords/eigenvalues-and-eigenvectors","display_name":"Eigenvalues and eigenvectors","score":0.45660001039505005},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.42890000343322754},{"id":"https://openalex.org/keywords/mahalanobis-distance","display_name":"Mahalanobis distance","score":0.40779998898506165},{"id":"https://openalex.org/keywords/mnist-database","display_name":"MNIST database","score":0.38920000195503235},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.3837999999523163},{"id":"https://openalex.org/keywords/dimension","display_name":"Dimension (graph theory)","score":0.35519999265670776}],"concepts":[{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5357999801635742},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.5091999769210815},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.4916999936103821},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4684000015258789},{"id":"https://openalex.org/C158693339","wikidata":"https://www.wikidata.org/wiki/Q190524","display_name":"Eigenvalues and eigenvectors","level":2,"score":0.45660001039505005},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.42890000343322754},{"id":"https://openalex.org/C1921717","wikidata":"https://www.wikidata.org/wiki/Q1334846","display_name":"Mahalanobis distance","level":2,"score":0.40779998898506165},{"id":"https://openalex.org/C190502265","wikidata":"https://www.wikidata.org/wiki/Q17069496","display_name":"MNIST database","level":3,"score":0.38920000195503235},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.3837999999523163},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3743000030517578},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.36489999294281006},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.35519999265670776},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.33469998836517334},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.3086000084877014},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.29260000586509705},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.275299996137619},{"id":"https://openalex.org/C129844170","wikidata":"https://www.wikidata.org/wiki/Q41299","display_name":"Quadratic equation","level":2,"score":0.2732999920845032},{"id":"https://openalex.org/C112633086","wikidata":"https://www.wikidata.org/wiki/Q381287","display_name":"White noise","level":2,"score":0.27090001106262207},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.2662000060081482},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.2623000144958496},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2614000141620636},{"id":"https://openalex.org/C121864883","wikidata":"https://www.wikidata.org/wiki/Q677916","display_name":"Statistical physics","level":1,"score":0.2606000006198883},{"id":"https://openalex.org/C98306654","wikidata":"https://www.wikidata.org/wiki/Q2409122","display_name":"Spectral theory","level":3,"score":0.25999999046325684},{"id":"https://openalex.org/C200331156","wikidata":"https://www.wikidata.org/wiki/Q506041","display_name":"Jacobian matrix and determinant","level":2,"score":0.2597000002861023}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.08764","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08764","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.08764","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08764","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[{"display_name":"Affordable and clean energy","score":0.4328843653202057,"id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Deep":[0],"vision":[1],"models":[2,141,179],"degrade":[3],"sharply":[4],"in":[5,9,193],"low-data":[6,194],"regimes,":[7],"particularly":[8],"medical":[10],"imaging":[11],"where":[12],"labeled":[13],"samples":[14],"are":[15,96],"scarce.":[16],"We":[17,47,196],"show":[18,81,164],"this":[19,112,132],"arises":[20],"not":[21],"merely":[22],"from":[23,26,71],"overfitting":[24],"but":[25],"a":[27,49,99,108,118,190,213],"geometric":[28],"failure:":[29],"finite-sample":[30,53],"noise":[31,89],"corrupts":[32],"the":[33,37,41,58,62,88,151],"embedding":[34],"covariance,":[35],"collapsing":[36],"eigengap":[38],"and":[39,77,129,149,160,172,201,207],"limiting":[40],"number":[42,63],"of":[43,52,64],"recoverable":[44,59],"signal-bearing":[45],"modes.":[46],"develop":[48],"spectral":[50,110,138,187,210],"theory":[51,76],"representation":[54],"learning":[55,135],"that":[56,66,82,103,145,165],"quantifies":[57],"dimension":[60],"K(N),":[61],"eigenmodes":[65],"can":[67,114],"be":[68,115],"stably":[69],"estimated":[70],"N":[72],"samples.":[73],"Using":[74],"perturbation":[75],"concentration":[78],"bounds,":[79],"we":[80,163],"only":[83],"modes":[84,171],"with":[85],"eigenvalues":[86],"above":[87],"floor":[90],"$\\|\\hat\u03a3":[91],"-":[92],"\u03a3\\|_{\\mathrm{op}}":[93],"\\sim":[94],"\\sqrt{D/N}$":[95],"reliable,":[97],"yielding":[98],"truncated":[100,119,198],"Mahalanobis":[101,199],"energy":[102,113,200],"governs":[104],"classification":[105],"performance.":[106],"Under":[107],"power-law":[109],"model,":[111],"approximated":[116],"by":[117],"Riemann":[120],"zeta":[121],"function,":[122],"linking":[123],"eigenvalue":[124],"decay":[125],"to":[126,203,216],"data":[127,156,218],"efficiency":[128],"AUC.":[130],"Within":[131],"framework,":[133],"multimodal":[134,166],"acts":[136],"as":[137,189,212],"stabilization:":[139],"vision-language":[140],"impose":[142],"low-rank":[143],"constraints":[144],"suppress":[146],"noise-dominated":[147],"directions":[148],"preserve":[150],"eigengap,":[152],"increasing":[153],"K(N)":[154,202],"under":[155],"scarcity.":[157],"Across":[158],"MNIST":[159],"multi-disease":[161],"neuroimaging,":[162],"training":[167],"maintains":[168],"more":[169],"stable":[170],"improves":[173],"class":[174],"separation,":[175],"even":[176],"when":[177],"unimodal":[178],"achieve":[180],"comparable":[181],"few-shot":[182],"accuracy.":[183],"These":[184],"results":[185],"identify":[186],"collapse":[188],"fundamental":[191],"bottleneck":[192],"learning.":[195],"use":[197],"diagnose":[204],"encoder":[205],"quality,":[206],"introduce":[208],"zeta-based":[209],"filtering":[211],"principled":[214],"approach":[215],"improve":[217],"efficiency.":[219]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-13T00:00:00"}