{"id":"https://openalex.org/W7127903874","doi":"https://doi.org/10.48550/arxiv.2602.05258","title":"CoPE: Clipped RoPE as A Scalable Free Lunch for Long Context LLMs","display_name":"CoPE: Clipped RoPE as A Scalable Free Lunch for Long Context LLMs","publication_year":2026,"publication_date":"2026-02-05","ids":{"openalex":"https://openalex.org/W7127903874","doi":"https://doi.org/10.48550/arxiv.2602.05258"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2602.05258","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.05258","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,"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2602.05258","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5125229407","display_name":"Haoran Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Haoran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048891976","display_name":"Sucheng Ren","orcid":"https://orcid.org/0000-0003-4730-8435"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ren, Sucheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125128549","display_name":"Alan Yuille","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuille, Alan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5125124166","display_name":"Feng Ryan Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Feng","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":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/T10028","display_name":"Topic Modeling","score":0.4074999988079071,"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":0.4074999988079071,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.20440000295639038,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.1266999989748001,"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/rope","display_name":"Rope","score":0.7498000264167786},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.652400016784668},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5796999931335449},{"id":"https://openalex.org/keywords/clipping","display_name":"Clipping (morphology)","score":0.5083000063896179},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.428600013256073},{"id":"https://openalex.org/keywords/outlier","display_name":"Outlier","score":0.38690000772476196},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.38179999589920044}],"concepts":[{"id":"https://openalex.org/C162269090","wikidata":"https://www.wikidata.org/wiki/Q1156047","display_name":"Rope","level":2,"score":0.7498000264167786},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6769999861717224},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.652400016784668},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5796999931335449},{"id":"https://openalex.org/C2776848632","wikidata":"https://www.wikidata.org/wiki/Q853463","display_name":"Clipping (morphology)","level":2,"score":0.5083000063896179},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46560001373291016},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.428600013256073},{"id":"https://openalex.org/C79337645","wikidata":"https://www.wikidata.org/wiki/Q779824","display_name":"Outlier","level":2,"score":0.38690000772476196},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.38179999589920044},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.38109999895095825},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.37470000982284546},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.37299999594688416},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.36419999599456787},{"id":"https://openalex.org/C2776190703","wikidata":"https://www.wikidata.org/wiki/Q488148","display_name":"Slicing","level":2,"score":0.3458999991416931},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.3239000141620636},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.30720001459121704}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2602.05258","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.05258","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,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2602.05258","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.05258","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,"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"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Rotary":[0],"Positional":[1],"Embedding":[2],"(RoPE)":[3],"is":[4],"a":[5,77,138],"key":[6],"component":[7],"of":[8,86],"context":[9,128],"scaling":[10],"in":[11],"Large":[12],"Language":[13],"Models":[14],"(LLMs).":[15],"While":[16],"various":[17],"methods":[18],"have":[19],"been":[20],"proposed":[21],"to":[22,25,44,117,126],"adapt":[23],"RoPE":[24,42,60,118],"longer":[26],"contexts,":[27],"their":[28],"guiding":[29],"principles":[30],"generally":[31],"fall":[32],"into":[33],"two":[34],"categories:":[35],"(1)":[36],"out-of-distribution":[37],"(OOD)":[38],"mitigation,":[39],"which":[40,52],"scales":[41],"frequencies":[43],"accommodate":[45],"unseen":[46],"positions,":[47],"and":[48,94,134,147],"(2)":[49],"Semantic":[50],"Modeling,":[51],"posits":[53],"that":[54,110,123],"the":[55],"attention":[56],"scores":[57],"computed":[58],"with":[59],"should":[61],"always":[62],"prioritize":[63],"semantically":[64],"similar":[65],"tokens.":[66],"In":[67],"this":[68],"work,":[69],"we":[70],"unify":[71],"these":[72],"seemingly":[73],"distinct":[74],"objectives":[75],"through":[76],"minimalist":[78],"intervention,":[79],"namely":[80],"CoPE:":[81],"soft":[82,114],"clipping":[83,115],"lowfrequency":[84],"components":[85],"RoPE.":[87],"CoPE":[88,136],"not":[89],"only":[90],"eliminates":[91],"OOD":[92],"outliers":[93],"refines":[95],"semantic":[96],"signals,":[97],"but":[98],"also":[99],"prevents":[100],"spectral":[101],"leakage":[102],"caused":[103],"by":[104],"hard":[105],"clipping.":[106],"Extensive":[107],"experiments":[108],"demonstrate":[109],"simply":[111],"applying":[112],"our":[113,131],"strategy":[116],"yields":[119],"significant":[120],"performance":[121],"gains":[122],"scale":[124],"up":[125],"256k":[127],"length,":[129],"validating":[130],"theoretical":[132],"analysis":[133],"establishing":[135],"as":[137],"new":[139],"state-of-the-art":[140],"for":[141],"length":[142],"generalization.":[143],"Our":[144],"code,":[145],"data,":[146],"models":[148],"are":[149],"available":[150],"at":[151],"https://github.com/hrlics/CoPE.":[152]},"counts_by_year":[],"updated_date":"2026-08-16T07:02:28.622633","created_date":"2026-02-07T00:00:00"}