{"id":"https://openalex.org/W4412945452","doi":"https://doi.org/10.18653/v1/2025.acl-long.581","title":"LongRecipe: Recipe for Efficient Long Context Generalization in Large Language Models","display_name":"LongRecipe: Recipe for Efficient Long Context Generalization in Large Language Models","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412945452","doi":"https://doi.org/10.18653/v1/2025.acl-long.581"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.581","doi":"https://doi.org/10.18653/v1/2025.acl-long.581","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.581","pdf_url":"https://aclanthology.org/2025.acl-long.581.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"bronze","oa_url":"https://aclanthology.org/2025.acl-long.581.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5076053801","display_name":"Zhiyuan Hu","orcid":"https://orcid.org/0000-0003-2212-0268","observed_orcids":["https://orcid.org/0000-0003-2212-0268","https://orcid.org/0000-0003-4697-9915","https://orcid.org/0009-0000-5535-500X"]},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Zhiyuan Hu","raw_affiliation_strings":["National University of Singapore ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National University of Singapore ,","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111226576","display_name":"Yuliang Liu","orcid":"https://orcid.org/0000-0001-7165-4341","observed_orcids":["https://orcid.org/0000-0001-7165-4341"]},"institutions":[{"id":"https://openalex.org/I4405279436","display_name":"Shanghai Innovation Institute","ror":"https://ror.org/01fatbw66","country_code":"CN","type":"education","lineage":["https://openalex.org/I4405279436"]},{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuliang Liu","raw_affiliation_strings":["Nanjing University ,","Shanghai Innovation Institute"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanjing University ,","institution_ids":["https://openalex.org/I881766915"]},{"raw_affiliation_string":"Shanghai Innovation Institute","institution_ids":["https://openalex.org/I4405279436"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036110657","display_name":"Jinman Zhao","orcid":"https://orcid.org/0009-0001-1511-8875","observed_orcids":["https://orcid.org/0009-0001-1511-8875"]},"institutions":[{"id":"https://openalex.org/I185261750","display_name":"University of Toronto","ror":"https://ror.org/03dbr7087","country_code":"CA","type":"education","lineage":["https://openalex.org/I185261750"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Jinman Zhao","raw_affiliation_strings":["University of Toronto ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Toronto ,","institution_ids":["https://openalex.org/I185261750"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073554817","display_name":"Suyuchen Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210164802","display_name":"Mila - Quebec Artificial Intelligence Institute","ror":"https://ror.org/05c22rx21","country_code":"CA","type":"facility","lineage":["https://openalex.org/I4210164802"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Suyuchen Wang","raw_affiliation_strings":["Mila , Qu\u00e9bec AI Institute /"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Mila , Qu\u00e9bec AI Institute /","institution_ids":["https://openalex.org/I4210164802"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110665331","display_name":"WangYan WangYan","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"WangYan WangYan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Wei Shen","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Shen","raw_affiliation_strings":["Nanjing University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanjing University ,","institution_ids":["https://openalex.org/I881766915"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Qing Gu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qing Gu","raw_affiliation_strings":["Nanjing University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanjing University ,","institution_ids":["https://openalex.org/I881766915"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Anh Tuan Luu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Anh Tuan Luu","raw_affiliation_strings":["Nanyang Technological University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanyang Technological University ,","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090171111","display_name":"See-Kiong Ng","orcid":"https://orcid.org/0000-0001-6565-7511","observed_orcids":["https://orcid.org/0000-0001-6565-7511"]},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"See-Kiong Ng","raw_affiliation_strings":["National University of Singapore ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National University of Singapore ,","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082553860","display_name":"Zhiwei Jiang","orcid":"https://orcid.org/0000-0001-5243-4992","observed_orcids":["https://orcid.org/0000-0001-5243-4992"]},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhiwei Jiang","raw_affiliation_strings":["Nanjing University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanjing University ,","institution_ids":["https://openalex.org/I881766915"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5065675832","display_name":"Bryan Hooi","orcid":"https://orcid.org/0000-0002-5645-1754","observed_orcids":["https://orcid.org/0000-0002-5645-1754"]},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]},{"id":"https://openalex.org/I70931966","display_name":"Universit\u00e9 de Montr\u00e9al","ror":"https://ror.org/0161xgx34","country_code":"CA","type":"education","lineage":["https://openalex.org/I70931966"]}],"countries":["CA","SG"],"is_corresponding":false,"raw_author_name":"Bryan Hooi","raw_affiliation_strings":["National University of Singapore ,","Universit\u00e9 de Montr\u00e9al ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National University of Singapore ,","institution_ids":["https://openalex.org/I165932596"]},{"raw_affiliation_string":"Universit\u00e9 de Montr\u00e9al ,","institution_ids":["https://openalex.org/I70931966"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":7,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.9496,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.71955189,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":93,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"11857","last_page":"11870"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9496999979019165,"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":0.9496999979019165,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.01720000058412552,"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":0.016300000250339508,"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/large-language-models","display_name":"large language models","score":0.8529999852180481},{"id":"https://openalex.org/keywords/long-text-generation","display_name":"long-form text generation","score":0.4620000123977661},{"id":"https://openalex.org/keywords/inference-efficiency","display_name":"inference efficiency","score":0.4729999899864197},{"id":"https://openalex.org/keywords/prompt-engineering","display_name":"prompt engineering","score":0.3050000071525574}],"concepts":[{"id":"https://openalex.org/C2778671685","wikidata":"https://www.wikidata.org/wiki/Q219239","display_name":"Recipe","level":2,"score":0.9250841736793518},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7157648205757141},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.6146450042724609},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6097399592399597},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4602218568325043},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42519354820251465},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.3657289743423462},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.15650206804275513},{"id":"https://openalex.org/C95457728","wikidata":"https://www.wikidata.org/wiki/Q309","display_name":"History","level":0,"score":0.07604324817657471},{"id":"https://openalex.org/C166957645","wikidata":"https://www.wikidata.org/wiki/Q23498","display_name":"Archaeology","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.581","doi":"https://doi.org/10.18653/v1/2025.acl-long.581","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.581","pdf_url":"https://aclanthology.org/2025.acl-long.581.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2409.00509","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2409.00509","pdf_url":"https://arxiv.org/pdf/2409.00509","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.2409.00509","doi":"https://doi.org/10.48550/arxiv.2409.00509","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2409.00509","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":"doi:10.18653/v1/2025.acl-long.581","doi":"https://doi.org/10.18653/v1/2025.acl-long.581","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.581","pdf_url":"https://aclanthology.org/2025.acl-long.581.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412945452.pdf","grobid_xml":"https://content.openalex.org/works/W4412945452.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W6903985660","https://openalex.org/W4399358068","https://openalex.org/W4412888777","https://openalex.org/W4415795411","https://openalex.org/W4386942555","https://openalex.org/W4403851694","https://openalex.org/W4416120555","https://openalex.org/W7161232054","https://openalex.org/W4387210426","https://openalex.org/W4390961274"],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"face":[4],"significant":[5],"challenges":[6],"in":[7,34,125],"handling":[8],"long-context":[9],"tasks":[10],"because":[11],"of":[12,55,81,87,100,134,150],"their":[13,23],"limited":[14],"effective":[15,131],"context":[16,32,53,103,132],"window":[17,33,54,104,133],"size":[18],"during":[19],"pretraining,":[20],"which":[21],"restricts":[22],"ability":[24],"to":[25,114,139,144],"generalize":[26],"over":[27,111],"extended":[28],"sequences.Meanwhile,":[29],"extending":[30,51],"the":[31,52,78,101,121],"LLMs":[35,88,136],"through":[36],"post-pretraining":[37],"is":[38],"highly":[39],"resource-intensive.To":[40],"address":[41],"this,":[42],"we":[43,128],"introduce":[44],"LongRecipe,":[45],"an":[46],"efficient":[47],"training":[48,65,73,109,152],"strategy":[49],"for":[50],"LLMs,":[56],"including":[57],"impactful":[58],"token":[59],"analysis,":[60],"position":[61],"index":[62],"transformation,":[63],"and":[64,75,106],"optimization":[66],"strategies.It":[67],"simulates":[68],"long-sequence":[69],"inputs":[70],"while":[71,96],"maintaining":[72],"efficiency":[74],"significantly":[76],"improves":[77],"model's":[79],"understanding":[80],"long-range":[82],"dependencies.Experiments":[83],"on":[84],"three":[85],"types":[86],"show":[89],"that":[90],"Lon-gRecipe":[91],"can":[92,129],"utilize":[93],"long":[94],"sequences":[95],"requiring":[97],"only":[98],"30%":[99],"target":[102],"size,":[105],"reduces":[107],"computational":[108],"resource":[110],"85%":[112],"compared":[113],"full":[115],"sequence":[116],"training.Furthermore,":[117],"LongRecipe":[118],"also":[119],"preserves":[120],"original":[122],"LLM's":[123],"capabilities":[124],"general":[126],"tasks.Ultimately,":[127],"extend":[130],"open-source":[135],"from":[137],"8k":[138],"128k,":[140],"achieving":[141],"performance":[142],"close":[143],"GPT-4":[145],"with":[146,157],"just":[147],"one":[148],"day":[149],"dedicated":[151],"using":[153],"a":[154],"single":[155],"GPU":[156],"80G":[158],"memory.":[159]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-10-29T09:56:32.521200","created_date":"2025-10-10T00:00:00"}