{"id":"https://openalex.org/W4403849929","doi":"https://doi.org/10.48550/arxiv.2407.14207","title":"Longhorn: State Space Models are Amortized Online Learners","display_name":"Longhorn: State Space Models are Amortized Online Learners","publication_year":2024,"publication_date":"2024-07-19","ids":{"openalex":"https://openalex.org/W4403849929","doi":"https://doi.org/10.48550/arxiv.2407.14207"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2407.14207","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2407.14207","pdf_url":"https://arxiv.org/pdf/2407.14207","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2407.14207","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100461655","display_name":"Bo Liu","orcid":"https://orcid.org/0000-0002-4250-6339","observed_orcids":["https://orcid.org/0000-0002-4250-6339","https://orcid.org/0000-0001-5209-9063"]},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Liu, Bo","raw_affiliation_strings":["The University of Texas at Austin , ; Helixon"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Texas at Austin , ; Helixon","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005518384","display_name":"Rui Wang","orcid":"https://orcid.org/0000-0002-6298-0716","observed_orcids":["https://orcid.org/0000-0002-6298-0716","https://orcid.org/0009-0007-4565-9190"]},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wang, Rui","raw_affiliation_strings":["The University of Texas at Austin , ; Helixon"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Texas at Austin , ; Helixon","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087616209","display_name":"Lemeng Wu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Lemeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111345651","display_name":"Yihao Feng","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Feng, Yihao","raw_affiliation_strings":["The University of Texas at Austin , ; Helixon"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Texas at Austin , ; Helixon","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001594330","display_name":"Peter Stone","orcid":"https://orcid.org/0000-0002-6795-420X","observed_orcids":["https://orcid.org/0000-0002-6795-420X"]},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Stone, Peter","raw_affiliation_strings":["The University of Texas at Austin , ; Helixon"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Texas at Austin , ; Helixon","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5057648803","display_name":"Qiang Liu","orcid":"https://orcid.org/0000-0003-1037-2185","observed_orcids":["https://orcid.org/0000-0003-1037-2185","https://orcid.org/0009-0001-8120-5539"]},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Liu, Qiang","raw_affiliation_strings":["The University of Texas at Austin , ; Helixon"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Texas at Austin , ; Helixon","institution_ids":["https://openalex.org/I86519309"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"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":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":0.6352999806404114,"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":0.6352999806404114,"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":0.09009999781847,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.03220000118017197,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/state-space-models","display_name":"state-space model","score":0.9279999732971191},{"id":"https://openalex.org/keywords/longhorns","display_name":"Longhorn","score":0.722000002861023},{"id":"https://openalex.org/keywords/e-learning","display_name":"online learning","score":0.8069999814033508},{"id":"https://openalex.org/keywords/sequence-modeling","display_name":"sequence modeling","score":0.699999988079071},{"id":"https://openalex.org/keywords/next-token-prediction","display_name":"next-token prediction","score":0.7160000205039978},{"id":"https://openalex.org/keywords/linear-decoding","display_name":"linear decoding","score":0.5450000166893005},{"id":"https://openalex.org/keywords/associative-recall","display_name":"associative recall","score":0.8320000171661377},{"id":"https://openalex.org/keywords/mamba","display_name":"Mamba","score":0.5329999923706055},{"id":"https://openalex.org/keywords/language-modeling","display_name":"language modeling","score":0.7770000100135803},{"id":"https://openalex.org/keywords/visual-task","display_name":"visual task","score":0.5929999947547913},{"id":"https://openalex.org/keywords/sample-efficiency","display_name":"sample efficiency","score":0.7979999780654907}],"concepts":[{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.5637659430503845},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5511438846588135},{"id":"https://openalex.org/C15046945","wikidata":"https://www.wikidata.org/wiki/Q205295","display_name":"Longhorn beetle","level":2,"score":0.5069742202758789},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.5022141933441162},{"id":"https://openalex.org/C72434380","wikidata":"https://www.wikidata.org/wiki/Q230930","display_name":"State space","level":2,"score":0.45939916372299194},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.30675008893013},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.23244112730026245},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.23191189765930176},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.08602824807167053},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.08549284934997559},{"id":"https://openalex.org/C90856448","wikidata":"https://www.wikidata.org/wiki/Q431","display_name":"Zoology","level":1,"score":0.06942883133888245}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2407.14207","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2407.14207","pdf_url":"https://arxiv.org/pdf/2407.14207","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2407.14207","doi":"https://doi.org/10.48550/arxiv.2407.14207","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2407.14207","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":"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2407.14207","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2407.14207","pdf_url":"https://arxiv.org/pdf/2407.14207","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4403849929.pdf","grobid_xml":"https://content.openalex.org/works/W4403849929.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W6947989101","https://openalex.org/W7168027269","https://openalex.org/W7203721960","https://openalex.org/W7166699971","https://openalex.org/W4415036217","https://openalex.org/W4313442864","https://openalex.org/W3209374680","https://openalex.org/W7104985747","https://openalex.org/W7118000905","https://openalex.org/W6910784736"],"abstract_inverted_index":{"Modern":[0],"large":[1],"language":[2,148],"models":[3,34],"are":[4,171],"built":[5],"on":[6,54,106,143],"sequence":[7,20,27,145],"modeling":[8,146],"via":[9],"next-token":[10],"prediction.":[11],"While":[12],"the":[13,16,71,120,125,140],"Transformer":[14],"remains":[15],"dominant":[17],"architecture":[18],"for":[19,80,123],"modeling,":[21,149],"its":[22],"quadratic":[23],"decoding":[24,42],"complexity":[25],"in":[26,159],"length":[28],"poses":[29],"a":[30,37,111,156],"major":[31],"limitation.":[32],"State-space":[33],"(SSMs)":[35],"present":[36],"competitive":[38],"alternative,":[39],"offering":[40],"linear":[41,55],"efficiency":[43,161],"while":[44],"maintaining":[45],"parallelism":[46],"during":[47,176],"training.":[48],"However,":[49],"most":[50],"existing":[51],"SSMs":[52,77],"rely":[53],"recurrence":[56],"designs":[57],"that":[58,134,170],"appear":[59],"somewhat":[60],"ad":[61],"hoc.":[62],"In":[63],"this":[64,107],"work,":[65],"we":[66,109],"explore":[67],"SSM":[68,88,114],"design":[69,89],"through":[70],"lens":[72],"of":[73],"online":[74,82,93,126],"learning,":[75],"conceptualizing":[76],"as":[78],"meta-modules":[79],"specific":[81],"learning":[83,94],"problems.":[84],"This":[85],"approach":[86],"links":[87],"to":[90,163,173],"formulating":[91],"precise":[92],"objectives,":[95],"with":[96],"state":[97],"transition":[98],"rules":[99],"derived":[100],"from":[101],"solving":[102,124],"these":[103],"objectives.":[104],"Based":[105],"insight,":[108],"introduce":[110],"novel":[112],"deep":[113],"architecture,":[115],"Longhorn,":[116],"whose":[117],"update":[118],"resembles":[119],"closed-form":[121],"solution":[122],"associative":[127],"recall":[128],"problem.":[129],"Our":[130],"experimental":[131],"results":[132],"show":[133],"Longhorn":[135,154],"outperforms":[136],"state-of-the-art":[137],"SSMs,":[138],"including":[139],"Mamba":[141],"model,":[142],"standard":[144],"benchmarks,":[147],"and":[150,165],"vision":[151],"tasks.":[152],"Specifically,":[153],"achieves":[155],"1.8x":[157],"improvement":[158],"sample":[160],"compared":[162],"Mamba,":[164],"can":[166],"extrapolate":[167],"over":[168],"contexts":[169],"up":[172],"16x":[174],"longer":[175],"inference.":[177]},"counts_by_year":[],"updated_date":"2026-10-13T10:08:12.033832","created_date":"2024-10-29T00:00:00"}