{"id":"https://openalex.org/W7202295370","doi":"https://doi.org/10.64898/2026.08.06.743174","title":"AdaGeneBudget: Cell-Adaptive Gene-Token Allocation for Efficient Single-Cell Foundation Models","display_name":"AdaGeneBudget: Cell-Adaptive Gene-Token Allocation for Efficient Single-Cell Foundation Models","publication_year":2026,"publication_date":"2026-08-12","ids":{"openalex":"https://openalex.org/W7202295370","doi":"https://doi.org/10.64898/2026.08.06.743174"},"language":"en","primary_location":{"id":"doi:10.64898/2026.08.06.743174","doi":"https://doi.org/10.64898/2026.08.06.743174","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.08.06.743174","pdf_url":null,"source":{"id":"https://openalex.org/S4306402567","display_name":"bioRxiv (Cold Spring Harbor Laboratory)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2750212522","host_organization_name":"Cold Spring Harbor Laboratory","host_organization_lineage":["https://openalex.org/I2750212522"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"bioRxiv","raw_type":"posted-content"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.64898/2026.08.06.743174","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100662968","display_name":"Dohee Kim","orcid":"https://orcid.org/0000-0002-7084-3631","observed_orcids":["https://orcid.org/0000-0002-7084-3631","https://orcid.org/0009-0006-9909-9128","https://orcid.org/0009-0007-7539-6512"]},"institutions":[{"id":"https://openalex.org/I138925566","display_name":"Ewha Womans University","ror":"https://ror.org/053fp5c05","country_code":"KR","type":"education","lineage":["https://openalex.org/I138925566"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Dohee Kim","raw_affiliation_strings":["Division of Artificial Intelligence & Software, Ewha Womans University, Seoul, 03760, Republic of Korea"],"raw_orcid":"https://orcid.org/0009-0007-7539-6512","affiliations":[{"raw_affiliation_string":"Division of Artificial Intelligence & Software, Ewha Womans University, Seoul, 03760, Republic of Korea","institution_ids":["https://openalex.org/I138925566"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5066765700","display_name":"Uiwon Hwang","orcid":"https://orcid.org/0000-0001-5054-2236","observed_orcids":["https://orcid.org/0000-0001-5054-2236"]},"institutions":[{"id":"https://openalex.org/I138925566","display_name":"Ewha Womans University","ror":"https://ror.org/053fp5c05","country_code":"KR","type":"education","lineage":["https://openalex.org/I138925566"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Uiwon Hwang","raw_affiliation_strings":["Division of Artificial Intelligence & Software, Ewha Womans University, Seoul, 03760, Republic of Korea"],"raw_orcid":"https://orcid.org/0000-0001-5054-2236","affiliations":[{"raw_affiliation_string":"Division of Artificial Intelligence & Software, Ewha Womans University, Seoul, 03760, Republic of Korea","institution_ids":["https://openalex.org/I138925566"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5066765700"],"corresponding_institution_ids":["https://openalex.org/I138925566"],"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/T11289","display_name":"Single-cell and spatial transcriptomics","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},"topics":[{"id":"https://openalex.org/T11289","display_name":"Single-cell and spatial transcriptomics","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10885","display_name":"Gene expression and cancer classification","score":9.999999747378752e-05,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10621","display_name":"Gene Regulatory Network Analysis","score":0.0,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/single-cell-foundation-models","display_name":"single-cell foundation models","score":0.9729999899864197},{"id":"https://openalex.org/keywords/highly-variable-genes","display_name":"highly variable genes","score":0.8489999771118164},{"id":"https://openalex.org/keywords/scgpt","display_name":"scGPT","score":0.8309999704360962},{"id":"https://openalex.org/keywords/geneformer","display_name":"Geneformer","score":0.9940000176429749},{"id":"https://openalex.org/keywords/gpu-memory-reduction","display_name":"GPU memory reduction","score":0.6740000247955322},{"id":"https://openalex.org/keywords/lineage-markers","display_name":"lineage markers","score":0.34700000286102295}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5878000259399414},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.5577999949455261},{"id":"https://openalex.org/C141603448","wikidata":"https://www.wikidata.org/wiki/Q134830","display_name":"Prefix","level":2,"score":0.5425999760627747},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5382999777793884},{"id":"https://openalex.org/C112313634","wikidata":"https://www.wikidata.org/wiki/Q7886648","display_name":"Complement (music)","level":5,"score":0.5267000198364258},{"id":"https://openalex.org/C34388435","wikidata":"https://www.wikidata.org/wiki/Q2267362","display_name":"Bounded function","level":2,"score":0.47769999504089355},{"id":"https://openalex.org/C4679612","wikidata":"https://www.wikidata.org/wiki/Q866298","display_name":"Aggregate (composite)","level":2,"score":0.46059998869895935},{"id":"https://openalex.org/C2776321320","wikidata":"https://www.wikidata.org/wiki/Q857525","display_name":"Annotation","level":2,"score":0.44679999351501465},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.3725000023841858},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.36629998683929443},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.35899999737739563},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.34610000252723694},{"id":"https://openalex.org/C149629883","wikidata":"https://www.wikidata.org/wiki/Q660926","display_name":"Fraction (chemistry)","level":2,"score":0.33980000019073486},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3321000039577484},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.3160000145435333},{"id":"https://openalex.org/C12426560","wikidata":"https://www.wikidata.org/wiki/Q189569","display_name":"Basis (linear algebra)","level":2,"score":0.30889999866485596},{"id":"https://openalex.org/C207467116","wikidata":"https://www.wikidata.org/wiki/Q4385666","display_name":"Inverse","level":2,"score":0.30809998512268066},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.2969000041484833},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2953000068664551},{"id":"https://openalex.org/C70721500","wikidata":"https://www.wikidata.org/wiki/Q177005","display_name":"Computational biology","level":1,"score":0.2913999855518341},{"id":"https://openalex.org/C102392041","wikidata":"https://www.wikidata.org/wiki/Q592860","display_name":"Sliding window protocol","level":3,"score":0.2802000045776367},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2797999978065491},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2605000138282776},{"id":"https://openalex.org/C29202148","wikidata":"https://www.wikidata.org/wiki/Q287260","display_name":"Resource allocation","level":2,"score":0.25540000200271606}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.64898/2026.08.06.743174","doi":"https://doi.org/10.64898/2026.08.06.743174","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.08.06.743174","pdf_url":null,"source":{"id":"https://openalex.org/S4306402567","display_name":"bioRxiv (Cold Spring Harbor Laboratory)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2750212522","host_organization_name":"Cold Spring Harbor Laboratory","host_organization_lineage":["https://openalex.org/I2750212522"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"bioRxiv","raw_type":"posted-content"}],"best_oa_location":{"id":"doi:10.64898/2026.08.06.743174","doi":"https://doi.org/10.64898/2026.08.06.743174","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.08.06.743174","pdf_url":null,"source":{"id":"https://openalex.org/S4306402567","display_name":"bioRxiv (Cold Spring Harbor Laboratory)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2750212522","host_organization_name":"Cold Spring Harbor Laboratory","host_organization_lineage":["https://openalex.org/I2750212522"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"bioRxiv","raw_type":"posted-content"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W1978394996","https://openalex.org/W2214074259","https://openalex.org/W2774307122","https://openalex.org/W2949177718","https://openalex.org/W2949681238","https://openalex.org/W3120511825","https://openalex.org/W3164692211","https://openalex.org/W4378838672","https://openalex.org/W4392168151","https://openalex.org/W4399387478","https://openalex.org/W4400249174","https://openalex.org/W4404793799","https://openalex.org/W4409484936"],"related_works":["https://openalex.org/W7196938282","https://openalex.org/W7168125956","https://openalex.org/W7169871858","https://openalex.org/W4416206635","https://openalex.org/W4387819650","https://openalex.org/W7167260123","https://openalex.org/W4417334374","https://openalex.org/W7160150119","https://openalex.org/W4414377853","https://openalex.org/W4405328867"],"abstract_inverted_index":{"Abstract":[0],"Single-cell":[1],"foundation":[2],"models":[3,142,282],"(scFMs)":[4],"represent":[5],"each":[6,60,83],"cell":[7,212],"using":[8,137],"sequences":[9],"of":[10,21,102],"gene-associated":[11],"tokens,":[12],"making":[13],"embedding":[14,240],"extraction":[15],"increasingly":[16],"costly":[17],"as":[18,250],"the":[19,93,103,125,238],"number":[20],"cells":[22],"and":[23,66,91,116,123,140,145,159,167,177,198,203,210,214,218,256],"expressed":[24],"genes":[25,38,63,221],"grows.":[26],"Existing":[27],"input":[28,34,272],"policies":[29],"typically":[30],"rely":[31],"on":[32,143],"fixed":[33],"budgets,":[35],"with":[36,86,149],"retained":[37,215],"determined":[39],"by":[40,114,185],"random":[41,202],"subsampling,":[42],"model-native":[43],"ranking,":[44],"or":[45],"a":[46,76,99,133,251,276],"dataset-level":[47],"highly":[48],"variable":[49],"gene":[50],"(HVG)":[51],"panel.":[52],"However,":[53],"they":[54],"do":[55],"not":[56],"jointly":[57],"determine,":[58],"for":[59,260,280],"cell,":[61],"which":[62],"to":[64,70,187,254,264],"retain":[65],"how":[67],"many":[68],"tokens":[69],"allocate.":[71],"We":[72,129],"introduce":[73],"AdaGeneBudget":[74,131,171,206,234],",":[75],"training-free":[77],"gene-token":[78,175,248],"selection":[79],"method":[80],"that":[81,97,270],"combines":[82],"gene\u2019s":[84],"expression":[85],"reference-derived":[87],"inverse":[88],"detection":[89],"frequency":[90],"retains":[92],"shortest":[94],"ranked":[95],"prefix":[96],"captures":[98],"target":[100],"fraction":[101],"cell\u2019s":[104],"expression-specificity":[105],"score":[106],"mass.":[107],"The":[108],"resulting":[109],"cell-specific":[110],"budget":[111],"is":[112],"bounded":[113],"minimum":[115],"maximum":[117],"lengths,":[118],"requires":[119],"no":[120],"cell-type":[121],"labels,":[122],"leaves":[124],"pretrained":[126,138],"backbone":[127],"unchanged.":[128],"evaluated":[130],"in":[132],"frozen-backbone":[134],"inference":[135],"setting":[136],"scGPT":[139,166],"Geneformer":[141,168],"Kang":[144],"PBMC":[146],"reference-mapping":[147],"tasks,":[148],"an":[150,160],"additional":[151],"scPRINT":[152],"comparison":[153],"against":[154],"its":[155],"official":[156],"HVG":[157,162,227],"policy":[158],"expressed-only":[161],"control.":[163],"Across":[164],"four":[165],"backbone\u2013dataset":[169],"pairs,":[170],"substantially":[172],"reduced":[173],"mean":[174],"counts":[176],"peak":[178],"GPU":[179],"memory":[180],"while":[181],"increasing":[182],"embedding-extraction":[183],"throughput":[184],"up":[186],"4.63\u00d7.":[188],"Despite":[189],"this":[190],"compression,":[191],"it":[192],"preserved":[193,208,237],"native-level":[194],"aggregate":[195],"annotation":[196,231],"utility":[197],"numerically":[199],"outperformed":[200],"token-matched":[201],"Top-Expression":[204],"baselines.":[205],"also":[207,268],"fine-grained":[209],"low-support":[211],"identities":[213],"lineage-marker":[216],"programs":[217],"stimulation-associated":[219],"pathway":[220],"under":[222,284],"compression.":[223],"In":[224],"scPRINT,":[225],"both":[226],"controls":[228],"achieved":[229],"higher":[230],"macro-F1,":[232],"whereas":[233],"more":[235],"faithfully":[236],"stimulation-induced":[239],"direction.":[241],"These":[242],"results":[243],"establish":[244],"biologically":[245],"informed,":[246],"cell-adaptive":[247,271],"allocation":[249,273],"practical":[252],"complement":[253],"architectural":[255],"systems-level":[257],"efficiency":[258],"methods":[259],"applying":[261],"existing":[262],"scFMs":[263],"new":[265],"datasets.":[266],"They":[267],"suggest":[269],"may":[274],"be":[275],"useful":[277],"design":[278],"principle":[279],"future":[281],"operating":[283],"finite":[285],"token":[286],"budgets.":[287]},"counts_by_year":[],"updated_date":"2026-11-09T13:22:03.289127","created_date":"2026-08-13T00:00:00"}