{"id":"https://openalex.org/W7224964379","doi":"https://doi.org/10.48550/arxiv.2610.39223","title":"Where Position Matters: Profiling-Guided Layer-Wise RoPE Allocation in Transformers","display_name":"Where Position Matters: Profiling-Guided Layer-Wise RoPE Allocation in Transformers","publication_year":2026,"publication_date":"2026-10-29","ids":{"openalex":"https://openalex.org/W7224964379","doi":"https://doi.org/10.48550/arxiv.2610.39223"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.39223","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.39223","pdf_url":"https://arxiv.org/pdf/2610.39223","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/2610.39223","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5160206328","display_name":"Zhiwei Liu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Zhiwei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101302285","display_name":"Boao Kong","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kong, Boao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5160162093","display_name":"Bing Liu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Bing","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":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":0.7973999977111816,"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":0.7973999977111816,"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":0.08810000121593475,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.03539999946951866,"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/rotary-position-embedding","display_name":"rotary position embedding","score":0.9850000143051147},{"id":"https://openalex.org/keywords/transformer-layers","display_name":"transformer layers","score":0.3659999966621399},{"id":"https://openalex.org/keywords/multi-head-latent-attention","display_name":"multi-head latent attention","score":0.7900000214576721},{"id":"https://openalex.org/keywords/score-calibration","display_name":"score calibration","score":0.593999981880188},{"id":"https://openalex.org/keywords/interlayer-interaction","display_name":"interlayer interaction","score":0.7879999876022339}],"concepts":[{"id":"https://openalex.org/C162269090","wikidata":"https://www.wikidata.org/wiki/Q1156047","display_name":"Rope","level":2,"score":0.8238999843597412},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.730400025844574},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6013000011444092},{"id":"https://openalex.org/C198082294","wikidata":"https://www.wikidata.org/wiki/Q3399648","display_name":"Position (finance)","level":2,"score":0.5565999746322632},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.5116000175476074},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.367000013589859},{"id":"https://openalex.org/C78780964","wikidata":"https://www.wikidata.org/wiki/Q7233193","display_name":"Position paper","level":2,"score":0.33079999685287476},{"id":"https://openalex.org/C2984019128","wikidata":"https://www.wikidata.org/wiki/Q1637433","display_name":"Physical modelling","level":2,"score":0.328900009393692},{"id":"https://openalex.org/C29202148","wikidata":"https://www.wikidata.org/wiki/Q287260","display_name":"Resource allocation","level":2,"score":0.320499986410141},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.31470000743865967},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.31200000643730164},{"id":"https://openalex.org/C42475967","wikidata":"https://www.wikidata.org/wiki/Q194292","display_name":"Operations research","level":1,"score":0.30809998512268066},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.2818000018596649},{"id":"https://openalex.org/C200601418","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability engineering","level":1,"score":0.27239999175071716},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2694000005722046},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.2615000009536743},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.2597000002861023},{"id":"https://openalex.org/C8505890","wikidata":"https://www.wikidata.org/wiki/Q605095","display_name":"Budget constraint","level":2,"score":0.2581999897956848},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.25769999623298645}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.39223","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.39223","pdf_url":"https://arxiv.org/pdf/2610.39223","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.2610.39223","doi":"https://doi.org/10.48550/arxiv.2610.39223","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.39223","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:2610.39223","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.39223","pdf_url":"https://arxiv.org/pdf/2610.39223","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/W7224964379.pdf","grobid_xml":"https://content.openalex.org/works/W7224964379.grobid-xml"},"referenced_works_count":14,"referenced_works":["https://openalex.org/W2973727699","https://openalex.org/W2988394319","https://openalex.org/W3196318247","https://openalex.org/W4384918448","https://openalex.org/W4389650671","https://openalex.org/W4392090070","https://openalex.org/W4396815229","https://openalex.org/W4401306886","https://openalex.org/W4403344704","https://openalex.org/W4405903187","https://openalex.org/W4414939355","https://openalex.org/W7133227460","https://openalex.org/W7135176124","https://openalex.org/W7208765302"],"related_works":["https://openalex.org/W7154222794","https://openalex.org/W7135176124","https://openalex.org/W7170143584","https://openalex.org/W4412724690","https://openalex.org/W4406892532","https://openalex.org/W6893319214","https://openalex.org/W7119165270","https://openalex.org/W7154841086","https://openalex.org/W7208765302","https://openalex.org/W7168241262"],"abstract_inverted_index":{"Rotary":[0],"Position":[1],"Embedding":[2],"(RoPE)":[3],"typically":[4],"uses":[5],"uniform":[6],"coverage":[7,23,38],"across":[8],"Transformer":[9],"layers":[10],"despite":[11],"heterogeneous":[12],"positional":[13,114],"requirements.":[14],"Phase-off":[15],"interventions":[16,24],"reveal":[17],"non-uniform,":[18],"model-specific":[19],"rotary-phase":[20],"dependence,":[21],"while":[22,100],"relate":[25],"it":[26],"to":[27,72,85,116],"coverage-upgrade":[28],"benefits.":[29],"We":[30],"propose":[31],"profiling-guided":[32],"layer-wise":[33],"RoPE":[34],"allocation,":[35],"assigning":[36],"high/low":[37],"under":[39],"a":[40],"fixed":[41],"budget":[42],"using":[43],"early":[44],"coverage-benefit":[45],"profiles.":[46],"The":[47],"framework":[48],"supports":[49],"fixed-width":[50],"query/key":[51],"representations":[52],"and":[53,83,88],"physical":[54,109],"rotary-dimension":[55],"allocation":[56],"in":[57],"Multi-head":[58],"Latent":[59],"Attention":[60],"(MLA).":[61],"Normalized-depth":[62],"mapping":[63],"transfers":[64],"profiles":[65],"within":[66],"model":[67],"families,":[68],"enabling":[69],"small-model":[70],"measurements":[71],"guide":[73],"larger-model":[74],"allocations":[75],"before":[76],"pretraining.":[77],"A":[78],"brief":[79],"analysis":[80],"relates":[81],"selection":[82],"transfer":[84],"score":[86],"calibration":[87],"inter-layer":[89],"interactions.":[90],"From-scratch":[91],"experiments":[92],"demonstrate":[93],"validation-loss":[94],"improvements":[95],"over":[96],"available":[97],"budget-matched":[98],"baselines,":[99],"analytical":[101],"resource":[102],"accounting":[103],"identifies":[104],"favorable":[105],"quality--storage--compute":[106],"trade-offs":[107],"for":[108],"MLA.":[110],"These":[111],"results":[112],"connect":[113],"demand":[115],"transferable,":[117],"efficient":[118],"architecture":[119],"design.":[120]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-31T00:00:00"}